Composite Structures 106 (2013) 570–580

Contents lists available at SciVerse ScienceDirect

Composite Structures
journal homepage: www.elsevier.com/locate/compstruct

Vibration and damping characteristics of hybrid carbon fiber composite

pyramidal truss sandwich panels with viscoelastic layers
Jinshui Yang, Jian Xiong, Li Ma ⇑, Bing Wang, Guoqi Zhang, Linzhi Wu
Center for Composite Materials, Harbin Institute of Technology, Harbin 150001, PR China

a r t i c l e i n f o a b s t r a c t

Article history: The vibration and damping performances of hybrid carbon fiber composite pyramidal truss sandwich
Available online 16 July 2013 panels with viscoelastic layers embedded in the face sheets were investigated in this paper. Hybrid car-
bon fiber composite pyramidal truss sandwich panels containing different thickness of viscoelastic layers
Keywords: were manufactured using a hot press molding method. Analytical models based on modal strain energy
Vibration damping approach were developed using ABAQUS software to estimate the damping property of the hybrid sand-
Composite wich structures. A set of modal tests were carried out to investigate the vibration and damping charac-
Sandwich panels
teristics of such hybrid sandwich panels with or without viscoelastic layers. The damping loss factors of
Pyramidal truss cores
Viscoelastic layer
composite slender beams with different fiber orientations were tested to determine the constitutive
damping properties of parent materials for such hybrid sandwich panels. The numerical simulation
results showed good agreement with the experimental tests. The damping loss factors of hybrid sand-
wich panels increased distinctly compared with previous sandwich panels due to the viscoelastic layer
embedded in the face sheets.
Ó 2013 Elsevier Ltd. All rights reserved.

1. Introduction sandwich materials mixed with a viscoelastic layers [20–22] can

Cellular sandwich structures with periodic lattice truss cores
have advantages of application in engineering load-bearing struc-
tures because of their high specific strength, stiffness and potential
of multi-functional applications [1–5]. Generally, lattice truss cores
have many kinds of topologies including tetrahedral [6,7], pyrami-
dal [8] and Kagome [9] configurations. Compared to metallic or al-
loy materials, using fiber-reinforce composites in sandwich
structures usually can obtain an efficient weight reduction without
strength and stiffness penalties. It needs to be pointed out that
most of research effort about the carbon fiber lattice truss core
sandwich structures focuses on the manufacturing techniques
and mechanical properties under compression, shear, bending
and impact loading [10–14]. However, to the authors’ knowledge,
the vibration and damping characteristics of lattice truss sandwich
structures have not been studied previously. As we all know, pre-
mature failure of critical components can be caused if the struc-
tures served under the resonant vibration. The problem of
reducing the amplitudes of the vibrations is of vital importance
for the structural design. Owing to the main contribution of the
viscoelastic character of polymeric composites, fiber-reinforced
composite materials commonly have 10–100 times the damping
of metals [15–19]. In addition, the hybrid structures of composite
⇑ Corresponding author. Tel.: +86 451 86402739; fax: +86 451 86402386.
E-mail address: mali@hit.edu.cn (L. Ma).
provide an effective way to achieve higher damping of the struc-
ture. The purpose of this article is to study the vibration and the
damping properties of carbon fiber composite pyramidal truss
sandwich panels and such panels embedded with viscoelastic
layers.
In the present study, the dynamic responses and damping
characteristics of hybrid carbon fiber composite pyramidal truss
sandwich panels were investigated. The analytical models based
on modal strain energy approach were proposed to estimate the
damping property of the structures in Section 2. Section 3, modal
experiments were performed to determine the constitutive damp-
ing properties of parent materials and the vibration and damping
characteristics of such hybrid sandwich panels. Section 4, the finite
element models were developed to predict the structural modal
parameters. The influences of different fiber orientations and thick-
ness of viscoelastic layers on the vibration property were investi-
gated and the conclusion would be drawn in the last.
2. Analytical models
2.1. Modal strain energy approach
The initial works on the damping analysis of fiber-reinforced
composite materials have been developed by Adams et al. [23].
The numerical analyzes considering the transverse shear effect
have shown good agreement with experimental results. Then, the

0263-8223/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.compstruct.2013.07.015
 J. Yang et al. / Composite Structures 106 (2013) 570–580 571

established damping model of FRP composites by Adams et al.
named modal strain energy approach was adopted by following
authors in order to develop the damping analysis of various com-
posite structures [24–26]. Maheri et al. [27] also demonstrated that
modal strain energy method was effective for analyzing the vibra-
tion damping of honeycomb structural panels with carbon fiber
composite face sheets. In the following section, the modal strain
energy approach was applied in a finite element formulation to
solve for the damping properties of the hybrid carbon fiber com-
posite pyramidal truss sandwich panels contained viscoelastic
layers.
The concept of this method is that the damping characteristics
of a structure can be defined by the ratio of the energy dissipated
to the energy stored during a stress cycle. For all-composite pyra-
midal truss core sandwich panels, the damping characteristics are
obviously anisotropic which is different with isotropic materials.
Hence, the total structural damping loss factor can be expressed as:
Pn
gij U kij
g ¼ Pk¼1
n k
ði; j ¼ 1; 2; 3Þ
k¼1 U ij
where gij and U kij are the damping loss factors of composite parent
material and strain energy components of the element k relative
to the stress component rij. As shown in Fig. 1, number 1 is the fiber
direction, 2 is transverse to this direction and 3 is the through-
thickness direction. Their relationships can be written as the follow-
ing equation:
Z nance through the modal tests. Then an arbitrary angle of the spec-
1 imen (for example, 45° specimen) can be applied to obtained
U kij ¼ rkij ekij dV k
2 parameter g12 by the modal test combing finite element analysis
Thus, the contributions of strain energy components corre-
sponding to stress components can be expressed as:
DU ij
SEij ¼ ði; j ¼ 1; 2; 3Þ
DU total
where DUij is the total strain energy component corresponding to
the stress component rij, DUtotal is the total strain energy of the sys-
tem. SE11, SE22 and SE33 are the contributions of tension–compres-
sion deformation in the 1, 2 and 3 direction of the fiber-
reinforced structures, respectively. SE12, SE13 and SE23 are the con-
tributions of shearing deformation in the planes (1, 2), (1, 3) and
(2, 3), respectively.
The procedure for evaluating damping of composite sandwich
structures was performed using a commercial finite element
software ABAQUS 6.10. The natural frequencies, mode shapes
and the properties (such as stress, strain and volume) of each
element can be obtained by the Lanczos eigensolver [28]. For the
anisotropic property of the composite structure, it was necessary
to further decompose the total strain energy into several different
components associated with different stress components. Unfortu-
nately, only the total strain energy outputs were shown in the
finite element analysis results. The solution procedure of the
damping loss factors of the model was shown in Fig. 2. The total
number of elements and the volume of each element are extracted
firstly. Correspondingly, various stress and strain components of
each element were exported to calculate and cumulate various
strain energy and dissipated energy components of the whole
model. Finally the damping loss factors of parent material were
used to obtain the damping loss factors of the model by the data
post-processing on the basis of modal strain energy approach.
2.2. Analytical models for damping properties of parent material
In order to predict the damping characteristics of the lattice
sandwich structure, it was necessary to get the damping character-
istics of its parent materials. Carbon fiber composite beams with
different fiber orientations were fabricated and the constitutive
damping data can be obtained through the modal test combining
numerical calculation based on modal strain energy approach.
The structural damping loss factor of the slender composite
ð1Þ beam can be expressed as:
Pn
gij U kij
g ¼ Pk¼1
n k
ði; j ¼ 1; 2Þ ð4Þ
k¼1 U ij
According to relevance theory [29], the damping loss factor
parameters g11 and g22 can be measured by subjecting a 0° and a
90° rectangular beam respectively to the flexural vibration reso-
ð2Þ
using the solution procedure of the damping loss factors of the
model mentioned in Fig. 2. As the specimen was unidirectional
carbon fiber composite which belongs to transversely isotropic
materials, then we have that, g22 = g33, g12 = g13, g23 has little effect
ð3Þ
on the modal behavior of such hybrid sandwich panels [30], and
can be assumed as the same value as g12.
2.3. Analytical models for damping properties of hybrid sandwich
panels
The constitutive damping data of parent material have been
obtained in above section and it were applied to evaluate the
damping properties of sandwich panels with pyramidal truss cores
and the hybrid sandwich panels embedded with different thick-
ness of viscoelastic layers. The damping loss factor of such struc-
tures can be defined as:
Pn Pm
k
g
k¼1 ij U ij þ p¼1 gv U pv
g¼ Pn k Pm p
ði; j ¼ 1; 2; 3Þ ð5Þ
k¼1 U ij þ p¼1 U v
where gv and U pv are the damping loss factor for viscoelastic mate-
rial and strain energy of element p.

2 1
y x
θ

Fig. 1. Definition of the fiber direction.

572 J. Yang et al. / Composite Structures 106 (2013) 570–580

Fig. 2. Solution procedure of the damping loss factors of the model.

The transverse shear strain energies played a major role in 3. Experiments

damping properties compared with in-plane strain energies, so
the damping performance of such structure was obtained by finite
element method considering the transverse shear effects and stress
effects through the thickness of laminate. Similar method was also
taken into account for such hybrid sandwich panels with viscoelas-
tic layers.
3.1. Materials
Carbon fiber composite slender beams with different fiber ori-
entations were fabricated from unidirectional carbon fiber/epoxy
prepregs of thickness 0.125 mm (T700/3234 composite, Beijing
 J. Yang et al. / Composite Structures 106 (2013) 570–580 573

Table 1 supplied by Shanghai Quiet New Material Technology Co., Ltd.

The material properties of unidirectional carbon/epoxy (T700/3234) laminate. The experimental results of Young’s modulus and damping loss
Symbol Value Property factor about this material were shown in Fig. 5.
E11 119 GPa Longitudinal stiffness
E22 8.7 GPa Transverse stiffness 3.2. Experimental process
E33 8.7 GPa Out-of-plane stiffness
t12, t13 0.32 Poisson’s ratio The test specimens of carbon fiber composite slender beams
t23 0.3 Poisson’s ratio
G12, G13 4 GPa Shear modulus
were supported horizontally as a cantilever beam in a fixture and
G23 3 GPa Shear modulus the first three bending modes of the structure were reported. The
q 1560 kg/m3 Density experimental equipment used in this work was shown in Fig. 6.
First of all, impact points were evenly arranged on the specimen,
and the force hammer was used to impact the points in proper or-
der after finished the relevant setup and the average values on the
Institute of Aeronautical Materials, China) using the hot press each point were obtained after impacting three times. Then signals
molding technique that were cured at 130 °C for 1.5 h under a gotten from force transducer (model SN30979, sensitivity
nominal pressure of 0.5 MPa on a mold plate. The detailed proper- 12.25 mV/N) and acceleration transducer (model SN46550, sensi-
ties of unidirectional carbon/epoxy composite laminate used in our tivity 10.07 mV/m/s2) were processed by the DEWETRON dynamic
experiments were listed in Table 1. There were three-groups of signal analyzers. As shown in Fig. 7, frequency response analysis
specimens and each group had three specimens which were was carried out to obtain the natural frequencies using the Fast
respectively processed into 0°, 45° and 90° rectangular beams. All Fourier transform method. The coherence function was applied
the specimens had a nominal length of 240 mm, a nominal width to make sure that the noise effect was small enough. The modal
of 20 mm and a nominal thickness of 2 mm could be tested in order circle fit was used to find the nearest resonance peak and offer
to get the damping properties of parent materials. an exact frequency damping estimation. The modal damping loss
Hybrid pyramidal truss sandwich panels embedded viscoelastic factor g relates to the damping ratios f by the relation g = 2f.
layers in the face sheets were fabricated using hot molding press Various modal shapes were visualized by the Frequency Response
method as described in the paper [10–13]. Both the top and bottom Function (FRF) Geometry.
face sheets of above hybrid sandwich structures were stacked by 8 Hybrid sandwich panels with and without viscoelastic layers
layers of carbon fiber/epoxy prepregs symmetrically cross-by [0°/ were arranged in the form of structure with one edge fixed and
90°/0°/90°]s. There were two unit cells along width direction and with the others free. The dimensions of the specimens were illus-
seven unit cells along length direction and the image of composite trated as following. The length a = 270 mm, the width b = 98 mm
sandwich panel and its unit cell were shown in Fig. 3. The visco- and the thickness h = 17 mm, the length of fixed area f = 15 mm,
elastic layers with different thickness highlighted by red color so the effective length of the structure a  f = 255 mm. The span
were inserted in the middle of the upper and lower facesheets as between the truss vertices k = 7 mm, the radius of truss member
shown in Fig. 4. The viscoelastic material used in this paper was d = 1.25 mm, the inclined angle x = 45° (Fig. 3(b)), the length of

Carbon fiber/epoxy
Carbon fiber/epoxy truss
(a) facesheets b
Fixed area filled
with wood
h

f a

(b)
2l cos ω + 2 k
k
2l cos ω + 2k
l

l sin ω
ω
d

Fig. 3. (a) Schematic illustration of pyramidal truss core sandwich panel with a boundary condition of cantilever. (b) Unit cell of pyramidal truss core.
574 J. Yang et al. / Composite Structures 106 (2013) 570–580

Layer 1 (0o)
Layer 2 (90o)
Layer 3 (0o)
Layer 4 (90o)
tf tv Viscoelastic layer
Layer 5 (90o)
Layer 6 (0o)
Layer 7 (90o)
Layer 8 (0o)

Fig. 4. Sandwich panels embedded with viscoelastic layers by the hot press molding technique.

0.30 mm, 0.45 mm, 0.60 mm and 0.75 mm (Fig. 4), respectively.
80 60
Different impact points and measuring points were arranged to de-
Young's modulus tect all vibration modes of the structure shown in Fig. 8. The first
Damping loss factor six modes were considered in the modal experiments.
70 50
Damping loss factor η (%)
Young's modulus (Mpa)

4. Simulation
60 40

The experimental results of carbon fiber composite slender

50 30 beams with different fiber orientations can be applied to the finite
element analysis to determine the damping loss factors of parent
materials for hybrid carbon fiber composite sandwich panels.
40 20 And these constitutive data (Table 2) were then used to predict
the relative damping characteristics of such hybrid sandwich
10
panels. The frequency extraction procedure was implemented in
30
0 200 400 600 800 1000 ABAQUS/Standard using the Lanczos eigensolver. A 3D solid model
Frequency (Hz) was developed to establish the structure, element C3D8R (8-node,
reduced integration) was used to mesh the model and a mesh con-
Fig. 5. The Young’s modulus and damping loss factor of the viscoelastic material. vergence study was carried out to ensure that the mesh refinement
in the sandwich panels was fine enough. The contribution of face
sheets and pyramidal truss cores on the damping characteristics
truss member l = 21.2 mm, the height of truss l sin x = 15 mm. In was firstly explored. Then the influences of different fiber orienta-
order to investigate the relationship between the thickness of vis- tions on the natural frequencies and damping loss factors were
coelastic layer and damping performance of such structures, the investigated. Finally, dynamic responses of such sandwich panels
thickness of viscoelastic layers tm were designed into 0.15 mm, were derived by using a mode superposition method.

Impact hammer

Accelerometer
Clamping block
Specimen

Signal conditioner Signal conditioner

Dynamic signal analyzer

Monitor

Fig. 6. The equipment for modal experiment.

 J. Yang et al. / Composite Structures 106 (2013) 570–580 575

(a) 100 2
52.442Hz
80 1

60
0

Amplitude (dB)

Coherence
40
-1
20
-2
0

-20 -3

-40 -4
0 200 400 600 800 1000
Frequency (Hz)

(b) Freq: 52.442 Hz

Damping ratio: 0.0022183
(c)

Fig. 7. (a) Frequency response analysis, (b) modal circle fit and (c) FRF geometry.

Impact points

Fixed area Measuring points

Fig. 8. Different impact points and measuring point arranged for the specimen.

5. Results and discussion
5.1. Damping characterization of parent materials
The first three natural frequencies and bending modal shapes of
the slender beams obtained by experiments and numerical calcula-
tions were shown in Figs. 9 and 10. A good agreement was ob-
served between the results derived from the experimental and
numerical results. It was reliable applying the experimental results
of damping loss factor parameters g11 and g22 to determine the
damping loss factor parameter g12 mentioned in Section 2.2.
Fig. 11 showed the experimental results obtained for damping in
576 J. Yang et al. / Composite Structures 106 (2013) 570–580

Table 2 3.0
The constitutive damping data of parent material for hybrid sandwich panels. o
θ=0
f (Hz) g11 (%) g22 = g33 (%) g12 = g13 (%) g23 (%) o
2.5 θ=45

Damping loss factor η i (%)

50 0.44 1.29 1.61 1.61 θ=90
o

300 0.55 1.63 2.00 2.00

600 0.67 2.05 2.46 2.46 2.0
1000 0.84 2.60 3.07 3.07

1.5

1200
o
θ=0 1.0
o
1000 θ=45
o
θ=90 0.5
Natural frequency f (Hz)

Experiment results
800
0 200 400 600 800 1000 1200 1400 1600
Frequency (Hz)
600
Fig. 11. The experimental results obtained for damping in the case of unidirectional
carbon fiber composite beams.
400

with different thickness of viscoelastic layers were given in Figs. 13

200
0 the boundary condition, joint damping and frictional damping
1 2
Mode
Fig. 9. The first three natural frequencies of the specimens obtained by the
experiments and numerical calculations.
the case of unidirectional carbon fiber composites. The results were
reported for the first three bending modes of the beams as function
of the frequency and three different fiber orientations (0°, 45°, 90°).
The solution procedure shown in Fig. 2 was implemented to obtain
the damping constitutive data of carbon fiber composite material.
The results were reported in Table 2 for the frequencies 50, 300,
600 and 1000 Hz. The vibration and damping performances of
the hybrid carbon fiber composite pyramidal truss sandwich
panels were summarized in the following section. The first six or-
der natural frequencies and modal shapes were deduced by the
Lanczos eigensolver. The corresponding damping loss factors were
obtained based on the modal strain energy approach.
5.2. Vibration and damping characterization
Fig. 12 showed the first six modal shapes of hybrid sandwich
panels deduced from experiments and finite element analysis.
The measured results showed a good agreement with the numeri-
cal predictions. Mode 1 was a transverse bending mode, mode 2
was a twisting mode and other modes were the results of their dif-
ferent superposition. The first six order natural frequencies and
modal damping loss factors of such hybrid sandwich panels were
investigated combining modal test and modeling prediction devel-
oped in Section 2. The results of fist six order natural frequencies
and damping loss factors for hybrid sandwich panels embedded
and 14. Compared to experiment results, the inevitable difference
for simulation prediction of the damping loss factors was due to
3
[15]. It was found that the damping loss factor of the sandwich
panels without viscoelastic layer in the range of 0.7–2% were much
higher than conventional materials and structures. Additionally,
the values of damping loss factor corresponds to the twist mode
were always higher than the transverse bending mode. The reason
was that greater shearing deformation could be induced by twist
mode to dissipate more energy. According to the modal strain en-
ergy approach, damping loss factor of such hybrid composite sand-
wich panels depended on the damping loss factor of its parent
materials and corresponding strain energy components. Fig. 15
showed the contribution of facesheets DUf/DUtotal and pyramidal
truss cores DUt/DUtotal on the damping characteristics of the total
structure. The results showed that the damping contribution of
facesheets played a more important role on the damping perfor-
mance of the structure than the pyramidal truss cores. With the in-
crease of the thickness of viscoelastic layer, damping loss factors of
the sandwich panels increased accordingly without significantly
changing its natural frequencies. Compared to the damping of
the structure without viscoelastic layer, the largest increase of
damping could be observed in such sandwich panels with
0.75 mm thickness viscoelastic layer that the damping loss factor
could reach 6.75%. Epoxy resin which penetrated into the visco-
elastic damping layers could improve the stiffness of the structure
to a certain degree during hot press molding fabrication, and the
modal natural frequencies slightly increased in high-order modes
with the increase of the thickness of viscoelastic layer.
5.3. The influence of fiber orientation
According to mechanics of composite materials, the bending–
stretching coupling and bending–twisting coupling existing in

Mode 1 Mode 2 Mode 3

Experiment

Simulation
Fig. 10. The first three modal shapes of the specimens obtained by the experiments and numerical calculations.
 J. Yang et al. / Composite Structures 106 (2013) 570–580 577

Mode 1

Mode 2

Mode 3

Mode 4

Mode 5

Mode 6
(a) Experiment results (b) Modelling results
Fig. 12. The first six modal shapes deduced from experiment and finite element analysis.

laminated plates could reduce the load-bearing capacity and natu-
ral frequency of structures, a symmetric angle-ply laminated plate
consisted of four ±h ply pairs through the thickness were consid-
ered in order to eliminate these effects for the damping properties
of hybrid sandwich structures. The increments of 15° of the fiber
orientation covering the 0–90° were implemented to study the
vibration and damping performances of such sandwich panels.
The variations of the damping loss factors and natural frequencies
as function of orientation were given in Figs. 16 and 17. Mode 1
was a transverse bending mode as mentioned above. In the case
of low-angle panels (0–30°), the longitudinal tension–compression
deformation SE11 that caused less energy loss was the sole contrib-
utor in such structure (Fig. 18) and the lower damping could also
be observed. On the contrary, the transverse tension–compression
578 J. Yang et al. / Composite Structures 106 (2013) 570–580

1200 4
Mode 1 (tν = 0 mm)
Mode 1 (tν = 0.15 mm)
1000 Mode 2 (tν = 0.15 mm)
Natural frequency f (Hz)

Damping loss factor η (%)

800

600 t ν = 0 mm 2
t ν = 0.15 mm
t ν = 0.3 mm
400 t ν = 0.45 mm
t ν = 0.6 mm 1
t ν = 0.75 mm
200 Experiment results

1 2 3 4 5 6
0
Mode 0 15 30 45 60 75 90
Fiber orientation θ (ο)
Fig. 13. Comparison of the natural frequencies obtained from modeling and
experimental investigation in the case of the structure with different thickness of
Fig. 16. Damping loss factors as function of the fiber orientation of structure with
viscoelastic layer.
viscoelastic layer of thickness 0.15 mm.

8
t ν = 0 mm 550
7 t ν = 0.15 mm Mode 1 (t ν = 0 mm)
500 Mode 1 (t ν = 0.15 mm)
t ν = 0.3 mm
6 t ν = 0.45 mm Mode 2 (t ν = 0.15 mm)
Damping loss factor η (%)

450
Natural frequency f (Hz)

t ν = 0.6 mm
5 t ν = 0.75 mm 400
Experiment results
4 350

300
3
250
2
200
1
150
0
1 2 3 4 5 6 100
0 15 30 45 60 75 90
Mode
Fiber orientation θ (o)
Fig. 14. Comparison of the modal damping loss factors obtained from modeling and
experimental investigation in the case of the structure with different thickness of Fig. 17. Natural frequencies as function of the fiber orientation of structure with
viscoelastic layer. viscoelastic layer of thickness 0.15 mm.
Contributions of strain energy components SE ij (%)

100
Facesheets SE 11
Truss cores SE 22
The contribution of the damping charateristics (%)

100 80 SE 33
SE 12
SE 13
60
80 SE 23

40
60

20
40

0
0 15 30 45 60 75 90
20
Fibre orientation θ (o)

Fig. 18. The contributions of strain energy components corresponding to stress

0 components for the mode 1 of structure without viscoelastic layer.
1 2 3 4 5 6
Mode

Fig. 15. The contribution of facesheets and pyramidal truss cores on the damping deformation SE22 and in-plane shearing deformation SE12 that
characteristics of the composite sandwich panel without viscoelastic layer. caused more energy loss played a major role in high-angle panels
 Contributions of strain energy components SE ij (%) J. Yang et al. / Composite Structures 106 (2013) 570–580 579

tν = 0 mm
100 60
tν = 0.15 mm
SE11 tν = 0.3 mm
SE22 50 tν = 0.45 mm
80 SE33 tν = 0.6 mm
tν = 075 mm
40

Amplitude (dB)
SE12
SE13
60
30
SE23

40 20

10
20
0

0 200 400 600 800 1000

0 15 30 45 60 75 90
Frequency (Hz)
Fibre orientation θ (o)
Fig. 21. Comparison of the dynamic responses of the structure with different
(a) Mode 1 thickness of viscoelastic layer.
Contributions of strain energy components SE ij (%)

100
SE11
significantly after interleaving a viscoelastic layer because of the
SE22 high damping property of the viscoelastic layer and the tension–
80 SE33 compression deformation in the thickness direction SE33 raised
SE12 apparently (Fig. 19). The values of damping for the mode 2 were
60
SE13 higher than the mode 1 because greater shearing deformation
SE23 was induced by twist mode to dissipate more energy.

40 5.4. Dynamic response

The vibrations of the structure were induced by an impulse

20
hammer and the acceleration transducer was used to acquire the
response of the structure. Different impact points and measuring
0 points (Fig. 8) were tested to investigate all the vibration modes
0 15 30 45 60 75 90
of such hybrid sandwich panels. The frequency range was from
Fibre orientation θ (o) 1 Hz to 1000 Hz in the dynamic experiments. Next, the steady-
(b) Mode 2 state dynamic responses of such structure were operated by finite
element analysis on the basis of modal superposition method. The
Fig. 19. The contributions of strain energy components corresponding to stress
modal damping loss factors obtained previously were used to
components for the structure with viscoelastic layer of thickness 0.15 mm.
study the vibration responses as functions of the frequency. The
experimental and numerical results of frequency responses were
compared with each other in Fig. 20. The agreement between each
60 Finite element analysis other was very good and minor difference can be found since the
Experimental results initial stress of impact force in the modal tests cannot be ensured
50 in the same level. Fig. 21 showed the dynamic responses of such
panels with different thickness of viscoelastic layer by experi-
40 ments. The amplitudes of the peaks significantly decreased with
Amplitude (dB)

the increase of the thickness of viscoelastic layers. The results

30 shown that present sandwich panels contained viscoelastic layers
could reduce the amplitude of vibration peak to 5–15 dB and it
20 could be used as lightweight and multifunctional benefits.

10 6. Conclusion

0 Experimental and numerical methods were carried out to study

200
Fig. 20. Comparison of the dynamic responses of the structure without viscoelastic
layer.
(30–90°). The damping was maximum at a fiber orientation of
about 60° for such sandwich panels under the combined effects
of all aspects. The similar phenomenon was also observed by
Berthelot et al. [26]. The damping of such panels increased
the vibration and damping performances of hybrid carbon fiber
400 600 800 1000 composite pyramidal truss sandwich panels containing viscoelastic
Frequency (Hz) layers. Such hybrid sandwich panels were fabricated by hot press
molding and different thickness of viscoelastic layers were embed-
ded in the middle of face sheets during the placement of carbon
fiber composite prepregs. The constitutive damping properties of
parent materials was obtained through the modal impact tests
combining the finite element analysis basing on modal strain
energy approach. The strategy of numerical simulation combining
data post-processing was used to estimate the damping loss factor
of such hybrid sandwich panels. The influences of facesheets with
580 J. Yang et al. / Composite Structures 106 (2013) 570–580
different fiber orientations on the damping and dynamic responses
of the structures have been also summarized. It was shown that
the modeling based on modal strain energy approach provided a
good way to estimation the damping characteristics of such hybrid
sandwich panels. The contribution of facesheets played a major
role on the damping of such structures compared with pyramidal
truss cores and it was a good way to embed the viscoelastic layers
into face sheets to control the vibration amplitude. The damping
loss factor of twist mode was always higher than the transverse
bending mode because of greater shearing deformation was in-
duced. Our results showed that insertion of viscoelastic layer in
the face sheets could distinctly increase its damping loss factor
without significantly changing its natural frequencies. This work
provided a way to study the damping characteristics of hybrid
composite sandwich panels with truss cores during multifunc-
tional applications.
Acknowledgment
This present work was supported by National Science Founda-
tion of China under Grant Nos. 11172080 and 11222216 and the
Fundamental Research Funds for the Central Universities under
Grant No. HIT.BRETIV.201301.
References
[1] Evans AG, Hutchinson JW, Fleck NA, Ashby MF, Wadley HNG. The topological
design of multifunctional cellular metals. Prog Mater Sci 2001;46:309–27.
[2] Wallach JC, Gibson LJ. Mechanical behavior of a three-dimensional truss
material. Int J Solids Struct 2001;38:7181–96.
[3] Vaziri A, Hutchinson JW. Metal sandwich plates subject to intense air shocks.
Int J Solids Struct 2007;44:2021–35.
[4] Ebrahimi H, Vaziri A. Metallic sandwich panels subjected to multiple intense
shocks. Int J Solids Struct 2013;50:1164–76.
[5] Wadley HNG. Multifunctional periodic cellular metals. Philos Trans R Soc A
2006;364:31–68.
[6] Deshpande VS, Ashby MF, Fleck NA. Foam topology bending versus stretching
dominated architectures. Acta Mater 2001;49(6):1035–40.
[7] Chiras S, Mumm DR, Evans AG. The structural performance of optimized truss
core panels. Int J Solids Struct 2002;39:4093–115.
[8] Cote F, Biagi R, Bart-Smith H, Deshpande VS. Structural response of pyramidal
core sandwich columns. Int J Solids Struct 2007;44:3533–56.
[9] Fan HL, Meng FH, Yang W. Sandwich panels with Kagome lattice cores
reinforced by carbon fibers. Compos Struct 2007;81:533–9.
[10] Wang B, Wu LZ, Ma L, Sun YG, Du SY. Mechanical behavior of the sandwich
structures with carbon fiber-reinforced pyramidal lattice truss core. Mater Des
2010;31:2659–63.
[11] Li M, Wu LZ, Ma L. Structural response of all-composite pyramidal truss core
sandwich columns in end compression. Compos Struct 2011;93:1964–72.
[12] Xiong J, Ma L, Pan SD, Wu LZ. Shear and bending performance of carbon fiber
composite sandwich panels with pyramidal truss cores. Acta Mater
2012;60:1455–66.
[13] Xiong J, Ma L, Wu LZ, Wang B. Fabrication and crushing behavior of low
density carbon fiber composite pyramidal truss structures. Compos Struct
2010;92:2695–702.
[14] Heimbs S, Cichosz J, Klaus M, Kilchert S, Johnson AF. Sandwich structures with
textile-reinforced composite foldcores under impact loads. Compos Struct
2010;92:1485–97.
[15] Chandra R, Singh SP, Gupta K. Damping studies in fiber-reinforced composites
– a review. Compos Struct 1999;46:41–51.
[16] Wang B, Yang M. Damping of honeycomb sandwich beams. Mater Process
Technol 2000;105:67–72.
[17] Maheri MR, Adams RD, Hugon J. Vibration damping in sandwich panels. J
Mater Sci 2008;43:6604–18.
[18] Fotsing ER, Sola M, Ross A, Ruiz E. Lightweight damping of composite
sandwich beams: experimental analysis. J Compos Mater 2012:1–11.
[19] Sarlin E, Liu Y. Vibration damping properties of steel/rubber/composite hybrid
structures. Compos Struct 2012;94:3327–35.
[20] Hajela P, Lin CY. Optimal design for viscoelastically damped beam structures.
Appl Mech Rev 1991;44:96–106.
[21] Bhimaraddi A. Sandwich beams theory and the analysis of constrained layer
damping. J Sound Vib 1995;179:591–602.
[22] Berthelot JM, Sefrani Y. Damping analysis of orthotropic composites with
interleaved viscoelastic layers: experimental investigation and discussion. J
Compos Mater 2006;40:1911–32.
[23] Ni RG, Adams RD. The damping and dynamic moduli of symmetric laminated
composite beams. Theoretical and experimental results. Compos Sci Technol
1984;18:104–21.
[24] Yim JH, Gillespie Jr JW. Damping characteristics of 0° and 90° AS4/3501-6
unidirectional laminates including the transverse shear effect. Compos Struct
2000;50:217–25.
[25] Berthelot JM, Sefrani Y. Damping analysis of unidirectional glass and Kevlar
fibre composites. Compos Sci Technol 2004;64:1261–78.
[26] Berthelot JM, Assarar M, Sefrani Y. Damping analysis of composite materials
and structures. Compos Struct 2008;85:189–204.
[27] Maheri MR, Adams RD, Hugon J. Vibration damping in sandwich panels. J
Mater Sci 2008;43:6604–18.
[28] ABAQUS. Standard user’s manual, Version 6.8. USA: Hibbitt, Karlsson and
Sorensen, Inc.; 2008.
[29] Lin DX, Ni R, Adams RD. Prediction and measurement of the vibrational
parameters of carbon and glass–fibre reinforced plastic plates. J Compos Mater
1984;18:132–52.
[30] Maheri MR, Adams RD. Finite-element prediction of modal response of
damped layered composite panels. Compos Sci Technol 1995;55:13–23.