Thin–Walled Structures 109 (2016) 174–184

Contents lists available at ScienceDirect

Thin–Walled Structures
journal homepage: www.elsevier.com/locate/tws

Full length article

Experimental and numerical investigation of dynamic plastic behavior of

tube with different thickness distribution under axial impact
crossmark
⁎
R. Rajabiehfarda, A. Darvizehb, , M. Alitavolia, H. Sadeghia, N. Noorzadehc, E. Maghdouric
a
Department of Mechanical Engineering, University of Guilan, P.O. Box 3756, Rasht, Iran
b
Department of Mechanical Engineering, Faculty of Engineering, Islamic Azad University, Anzali Branch, Bandar Anzali, Iran
c
Department of Mechanical Engineering, Ahrar Institute of Technology & Higher Education, Rasht, Iran

A R T I C L E I N F O A BS T RAC T

Keywords: In this study, the dynamic behavior of uniform thickness, stepped thickness and functionally graded thickness
Tube tubes under axial impact are investigated experimentally and numerically. A striking mass has been used to
Different thickness distribution impact tubes. Experimental tests are performed by using gas gun and numerical results are obtained by FE
Axial impact simulation. The effect of thickness distribution on shortening, energy absorption, axial force and buckling shape
Energy absorption
of tubes are investigated. It is found that a change in thickness distribution of tube can convert the buckling
Progressive buckling
shape from buckling with mild folds to progressive buckling and vice versa. In addition, it is found that stepped
thickness tube can be an approximate of functionally graded thickness tube which in this case almost, their
behavior is identical. This study reveals that stepped thickness and functionally graded thickness tubes in
comparison with uniform thickness tube absorb the same energy with more shortening and less peak load or
less mean load; thus, they are better energy absorption specimens. With comparing experimental and numerical
results, it is found that there is a good agreement between them.

1. Introduction aluminum tubes absorb the energy with folding mechanism.

Karagiozova and Jones [3] explored the influence of approximation
Tubes have been becoming increasingly popular in different in- of strain hardening modulus on the type of buckling shape (plastic or
dustries due to their economic, lightness and efficiency. Because of progressive buckling) and also the effect of axial inertia on initiation
their high energy absorption capacity, long crushing distance and high and development of buckling shape of tubes under axial impact. In the
ratio of energy absorption into weight, one of their most important dynamic plastic buckling "the whole length of a tube wrinkles before
applications is in the energy absorption systems. Thus, investigation on the large radial displacements develop [3]", and in the dynamic
dynamic behavior of tubes is significant. This paper aims to study the progressive buckling "the folds in a tube form consecutively [3]".
dynamic behavior of uniform thickness, stepped thickness and func- Energy absorption characteristics of tubes under axial loading were
tionally graded thickness tubes under axial impact. studied using nonlinear finite element software LS-DYNA by Tai et al.
The dynamic behavior of uniform thickness tubes was investigated [4]. This study revealed that with assuming of equal cross sectional for
experimentally [1], numerically [2–5] and theoretically [6–8] by two different materials, the energy absorption of tube made of high
different researchers. Florence and Goodier [1] experimentally inves- strength steel is higher than the energy absorption of tube made of mild
tigated the plastic buckling of tubes subjected to axial impact. In this steel. With investigation the effect of different parameters such as
paper, results of shortening, impact duration, half-wave number and material properties, geometry and impact velocity on the peak load of
buckling shape of tubes were reported. the tube which was subjected to axial impact, an approximate equation
Finite element analysis was used to investigate the buckling of tubes for estimation of peak load at a relatively low velocity (v0 < 40 (m/s))
subjected to axial impact [2–5]. By studying the effects of material was proposed by Chen and Ushijima [5].
properties, shell geometry, boundary and loading conditions on energy Lepik [6] studied the buckling improvement of elastic-plastic tubes
absorption and buckling shape of tubes by Karagiozova and Jones [2], under axial impact with Galerkin method. Axial impact of elastic-
it was found that a large proportion of the initial kinetic energy is plastic tubes with linear strain hardening material was studied using a
absorbed through a shortening mechanism in steel tubes, whereas discrete model by Karagiozova and Jones [7]. It was found that inertia

⁎
Corresponding author.
E-mail addresses: [email protected], [email protected] (A. Darvizeh).

http://dx.doi.org/10.1016/j.tws.2016.08.017
Received 25 February 2016; Received in revised form 23 August 2016; Accepted 24 August 2016
Available online 01 October 2016
0263-8231/ © 2016 Published by Elsevier Ltd.
R. Rajabiehfard et al. Thin–Walled Structures 109 (2016) 174–184

Fig. 1. Thickness, radius and length of U.Th, S.Th and FG.Th tubes (All dimensions are in millimeters (mm)).

properties of striking mass and shell geometry strongly affect the

buckling shape. The plastic buckling of tubes under axial impact was
investigated theoretically by Wang and Tian [8]. In this study, two
types of loading were investigated and the results of shortening, time
duration, half-wave number and buckling shape of tubes were reported.
The axial crushing of grooved tubes was investigated by different
researchers [9–13]. Mirmohammadsadeghi et al. [11] numerically and
experimentally investigated the axial crushing of grooved tubes. In this
study, the results of energy absorption, mean crushing load and initial
buckling load of grooved tubes with different geometric dimensions
were presented. It revealed that the crushing behavior and energy
absorption of grooved tubes can be controlled by different groove
distances. The effect of low density, low strength polyurethane foam on
the energy absorption of circumferentially grooved tubes was investi-
gated analytically and experimentally by Darvizeh et al. [13]. This study
revealed that circumferentially grooves prevent Euler buckling. Also, it
was found that used foam increase specific energy absorption approxi-
mately twice that of the empty tubes.
For improving of energy absorption properties and reduction of
initial peak loads, sinusoidal patterns corrugated tubes were used.
Buckling modes and energy absorption properties of sinusoidal pat-
terns corrugated tubes under axial impact were investigated by Liu
et al. [14]. In addition, the numerical analysis of aluminum foam filled
corrugated single and double circular tubes under axial impact was
investigated [15]. Fig. 3. Stationary tube and striking mass (a) tube on the rigid wall, (b) striking mass.
Moradpour et al. [16] performed a study on axial crushing behavior
of aluminum and mild steel circular tubes with circular holes on their in both of aluminum and mild steel circular tubes, symmetrical crushed
surface. The effect of rows number, holes number and hole diameters tubes and favorable performance will be obtained when rows number is
were investigated on axial crushing behavior of tubes. It was found that 5, holes number in each row is 12 and the diameter of each hole is

Fig. 2. Schematic diagram of experimental apparatus.

175
R. Rajabiehfard et al. Thin–Walled Structures 109 (2016) 174–184

Fig. 4. A view of longitudinal section at the moment that stationary tube is impacted by striking mass in FE simulation.

has not been investigated experimentally. So, in this paper, the

dynamic behavior of uniform thickness, stepped thickness and func-
tionally graded thickness tubes under axial impact are investigated
experimentally and numerically. Tubes are impacted by a striking
mass. Experimental tests are performed by using gas gun; also, FE
simulation was used to obtain numerical results. The effect of thickness
distribution on shortening, energy absorption, axial force and buckling
shape of tubes are investigated.

2. Uniform thickness, stepped thickness and functionally

graded thickness tubes

The thickness, radius and length of each uniform thickness, stepped

thickness and functionally graded thickness tubes are shown in Fig. 1.
Fig. 5. True stress-true plastic strain curve. It should be noted that U.Th, S.Th and FG.Th refer to uniform

Table 1
Shortening for U.Th, S.Th and FG.Th tubes under axial impact.

Velocity (m/s) Shortening (mm)

U.Th S.Th FG.Th

Numerical Experimental Numerical Experimental Numerical Experimental

30.12 1.89 1.98 2.44 2.17 4.63 4.25

41.69 3.48 2 6.14 5.43 7.24 6.8
50.93 5.33 3.5 10.33 9.43 10.6 9.17
58.11 7.27 5.23 14.23 12 12.2 14.2
63.41 9.04 7.7 15.78 12.1 13.96 14.3

6(mm). thickness, stepped thickness and functionally graded thickness tubes,

Tapered tubes with graded thickness under axial impact were respectively. The approximate weight of all used tubes is 68 or 70 (gr).
studied numerically and experimentally by Zhang et al. [17].
Numerical results were obtained using nonlinear finite element code 3. Experimental procedure
LS-DYNA. It was found that energy absorption properties of tapered
tubes are considerably higher than traditional tubes. Experimental results of tubes subjected to axial impact are obtained
Functionally graded thickness and tapered tubes under oblique by gas gun in Ahrar institute of Technology and Higher Education.
impact loading were investigated using FE simulation by Li et al. [18]. When an impact test is carried out, the load is measured by load
Energy absorption properties of functionally graded thickness, tapered cell. Variation of voltage caused by impact is measured by strain gauges
uniform thickness and straight uniform thickness tubes under oblique of load cell; an amplifier is used to amplify the measured voltage versus
loading were compared and the results presented that functionally time and then a digital storage oscilloscope records it. Finally, voltage-
graded thickness tubes show better performance as an energy absorber. time response is converted to force time response by a scale factor of
Alavi Nia et al. [19] studied axial quasi-static behavior of tubes with 1 mv=63.183 N. A schematic diagram of the experimental apparatus
various cross sections (circular, square, rectangular, hexagonal, trian- for load measuring is depicted in Fig. 2.
gular, pyramidal and conical) numerically and experimentally. It was A projectile with 372 (gr) mass impacts the stationary tube from the
found that circular tube has the most energy absorption capacity and end with less cross-sectional area axially; the stationary tube is placed
the most average force among all investigated sections. on the rigid wall as presented in Fig. 3(a). The striking mass as shown
Gumruk [20] used finite element program LS-DYNA3D to investi- in Fig. 3(b) is axially shot by gas gun.
gate the high velocity impact of aluminum tubes with axial linear
variable thickness, discontinuity and conical shaped, it was found that
4. Numerical simulation
a small change in tube geometry can convert the plastic and progressive
buckling.
Numerical simulation of U.Th, S.Th and FG.Th tubes under axial
In the previous papers, the effect of different thickness distribution
impact is done by FE simulation. In order to simulate the impact
on dynamic behavior of tubes subjected to axial high velocity impact
process, the wall which stationary tube is located on, and the striking

176
R. Rajabiehfard et al. Thin–Walled Structures 109 (2016) 174–184

Fig. 6. Experimental and numerical buckling shapes of U.Th tubes under axial impact with different velocities.

Fig. 7. Experimental and numerical buckling shapes of S.Th tube under axial impact with different velocities.

mass are simulated rigidly. The type of element used in the mesh is S.Th and FG.Th tubes subjected to axial impact are presented in
C3D8R. Also, for independence of numerical results from mesh and different velocities experimentally and numerically. The numerical
convergence of results, mesh size is considered 0.00035. According to results are validated with the aid of experimental results. Then,
mentioned mesh size, the number of mesh elements for shell is numerical energy absorption results are presented for three mentioned
obtained 158,760. Contact between all surfaces is considered in general tubes in different velocities. All used tubes are made of steel. The
contact. A view of longitudinal section at the moment that stationary mechanical properties of used steel tubes are determined according to
tube is impacted by striking mass in FE simulation is depicted in Fig. 4. metal tensile testing standard ASTM E8 using tensile testing machine
Santam in Islamic Azad university of Lashtenesha-Zibakenar branch.
5. Results and discussion The curve of true stress-true plastic strain for used material is depicted
in Fig. 5. The density, elastic modulus and poisson ratio are 7939 (m3/
Shortenings, buckling shapes and axial force-time curves of U.Th, kg), 200 (GPa) and 0.3, respectively. Cowper-Symond relation is used

177
R. Rajabiehfard et al. Thin–Walled Structures 109 (2016) 174–184

Fig. 8. Experimental and numerical buckling shapes of FG.Th tube under axial impact with different velocities.

Fig. 9. Experimental buckling shapes of U.Th, S.Th and FG.Th tubes at different velocities.

to apply strain rate effect in finite element analysis. The Cowper- Yd

1
⎛ ε̇ ⎞q
Symond relation is =1 + ⎜ ⎟
Ys ⎝D⎠ (1)

where Yd and Ys are dynamic and static yield stress, respectively, and ε̇

178
R. Rajabiehfard et al. Thin–Walled Structures 109 (2016) 174–184

Fig. 10. Comparing of the experimental and numerical force-time curves for U.Th, S.Th and FG.Th tubes at velocities 30.12, 41.69, 50.93, 58.11 and 63.41 (m/s).

179
R. Rajabiehfard et al. Thin–Walled Structures 109 (2016) 174–184

Fig. 11. Numerical buckling shapes together with axial force time curve of U.Th tube under axial impact at velocity 63.41 (m/s).

is strain rate. Also, D and q are constant coefficients of Cowper-Symond absorbed by increasing radial displacements of middle cross section,
relation, and their amount for steel material is 40.4 and 5, respectively and in FG.Th tube is absorbed by increasing radial displacements of
[21]. It should be noted that the striking mass is made of VCN150 second wrinkle.
(AISI4340) steel, and its outer surface is covered by Ertalon in order to With comparing of shortenings for three different types of tubes, it
convenience shooting. is found that the amounts of shortening for S.Th and FG.Th tubes are
close (except for velocity 30.12 (m/s)), but it's different from the
5.1. Shortening and buckling shape of U.Th, S.Th and FG.Th tubes amounts of shortening for U.Th tube. It should be noted that with
assuming equal equivalent cross sectional area, S.Th tube is an
Experimental and numerical shortening results and buckling approximate of FG.Th tube. Higher number of steps leads to more
shapes for three types of mentioned tubes under axial impact with compliance of S.Th tubes with FG.Th tubes. Because equivalent cross
different velocities are presented in Table 1 and Figs. (6–8), respec- sectional area of FG.Th tube is approximately equal with cross sectional
tively. With comparing experimental and numerical shortening results area of S.Th tube, therefore, a little difference between shortening
from Table 1 with each other and comparing experimental and results for S.Th and FG.Th tubes is reasonable. Parts with smaller cross
numerical buckling shapes according to Figs. (6–8) with each other, sectional areas of FG.Th tube in comparison with S.Th tube is involved
it is found that there is a good agreement between them. It should be in impact process at velocity 30.12 (m/s). So, at velocity 30.12 (m/s)
noted that incompatibility between some of experimental and numer- for FG.Th tube in comparison with S.Th tube larger shortening is
ical buckling shapes in Fig. 7, has been created for reasons, such as lack obtained. Also, from the presented results for shortening, it is known
of sufficient precision in preparing tubes. that shortening amounts of S.Th and FG.Th tubes are higher than U.Th
tube ones. Impaction of lower cross section of mentioned tubes leads to
higher shortenings for S.Th and FG.Th tubes. In other words, axial
5.1.1. Shortening of U.Th, S.Th and FG.Th tubes
shortening and radial displacement happens easier in lower cross
As it is clear from Table 1 and Figs. (6–8), shortenings for U.Th,
section.
S.Th and FG.Th tubes under axial impact show an increasing with
increasing of velocity. This is accordance with the fact that for tubes
under axial impact, a greater velocity causes a greater shortening. With 5.1.2. Buckling shape of U.Th, S.Th and FG.Th tubes
more investigation on presented shortening results and buckling As previously mentioned, buckling shapes of U.Th, FG.Th and S.Th
shapes for S.Th and FG.Th tubes in Table 1 and Figs. (7 and 8), tubes under axial impact with different velocities are presented in Figs.
respectively, it is found that shortening in velocities 58.11 and (6–8), respectively. Fig. 6, demonstrates that for U.Th tube at velocities
63.41 (m/s) is approximately identical for each of S.Th and FG.Th 30.12, 41.69, 50.93 and 58.11 (m/s), dynamic buckling with mild folds
tubes. The reason can be explained by using Figs. 7 and 8. As it is occurred, because total length of tubes are affected and two mild folds
demonstrated in these figures, more impact energy of velocity are formed at two ends of tubes. But for velocity 63.41 (m/s), dynamic
63.41 (m/s) in comparison with velocity 58.11 (m/s) in S.Th tube is progressive buckling occurred.

180
R. Rajabiehfard et al. Thin–Walled Structures 109 (2016) 174–184

Fig. 12. Numerical buckling shapes together with axial force time curve of S.Th tube under axial impact at velocity 63.41 (m/s).

Fig. 13. Numerical buckling shapes together with axial force time curve of FG.Th tube under axial impact at velocity 63.41 (m/s).

181
R. Rajabiehfard et al. Thin–Walled Structures 109 (2016) 174–184

Table 2 In addition, investigation on buckling shapes for FG.Th tubes that

Impact energy and energy absorption for U.Th, S.Th and FG.Th tubes under axial impact. are presented in Fig. 8 shows that for all velocities, dynamic progres-
sive buckling occurred. For velocities 30.12, 41.69 and 50.93 (m/s),
Velocity (m/s) Impact energy (J) Energy absorption (J)
dynamic progressive buckling with one fold occurred, but for velocities
U.Th S.Th FG.Th 58.11 and 63.41 (m/s), dynamic progressive buckling with two folds
occurred.
30.12 168.74 163.91 167.04 167.64
According to presented buckling shapes in Fig. 9, it is found with
41.69 323.28 319.44 322.11 322.08
50.93 482.46 479.66 480.82 480.66 changing of thickness distribution and constant remaining of other
58.11 628.08 625.68 626.58 626.16 parameters in tubes, buckling shapes can be transformed from
63.41 747.87 744.67 745.65 745.97 dynamic buckling with mild folds to dynamic progressive buckling
and vice versa. For example, buckling shape of U.Th and S.Th tubes at
velocity 30.12 (m/s) is dynamic buckling with mild folds, but buckling
shape of FG.Th tube at the same velocity is dynamic progressive
buckling. Also, buckling shapes of U.Th tube at velocities 41.69, 50.93
and 58.11 (m/s) are dynamic buckling with mild folds, but buckling
shapes of FG.Th and S.Th tubes at these velocities are progressive.
As it has explained before, S.Th tube is an approximate of FG.Th
tube. With comparing of buckling shapes of S.Th and FG.Th tubes with
each other, it is found that their buckling shapes are close to each other
except for velocity 30.12 (m/s), that as it was found before, this
incompatibility for shortening is also true. As mentioned before, at
velocity 30.12 (m/s), parts with smaller cross-sectional area from
FG.Th tube in comparison with S.Th tube is involved in impact process.
So, at velocity 30.12 (m/s) for FG.Th tube in comparison with S.Th
tube, larger radial displacements are obtained. According to Fig. 9, at
velocities 41.69 and 50.93 (m/s), dynamic progressive buckling with
one fold occurred for both S.Th and FG.Th tubes, and at velocities
58.11 and 63.41 (m/s) dynamic progressive buckling with two folds
occurred.

5.2. Axial load-time curves of U.Th, S.Th and FG.Th tubes

Experimental axial force-time curves together with buckling shapes

at different velocities for U.Th, S.Th and FG.Th tubes under axial
impact are depicted in Fig. 10. First peak load happens for U.Th, S.Th
and FG.Th tubes for all velocities, at a time range of 0–100 (μs). This
peak load creates one of the folds that is shown in Fig. 10. Also, it is
revealed that at investigated velocity range in this paper, the magnitude
of first peak load has not been affected by velocity, approximately.
Experimental axial force-time curves for the same velocities and
different geometries are given in Fig. 10. It is found out that peak loads
of U.Th tubes are higher than those of S.Th and FG.Th tubes at all
velocities. This is because, U.Th tube has a greater cross sectional area
in comparison with S.Th and FG.Th tubes. Greater cross section needs
a greater axial load to form folds. Thus, greater peak loads of U.Th tube
compared to peak loads of S.Th and FG.Th tubes is reasonable.
As mentioned before, S.Th tube is an approximate of FG.Th tube.
Therefore, according to Fig. 10, the behavior of axial load-time curves
of S.Th and FG.Th tubes at all velocities is approximately identical. In
addition, this figure demonstrates that first peak loads of S.Th tube is
greater or equal to first peak loads of FG.Th tube. This is because, the
first fold in S.Th tube happens in greater cross section and so, greater
axial load is required.
According to Fig. 10, the experimental and numerical force-time
curves for U.Th, S.Th and FG.Th tubes at velocities 30.12, 41.69, 50.93,
58.11 and 63.41 (m/s) are compared. As it is clear from this figure,
Fig. 14. Energy absorption versus shortening for the same thickness distribution and
different velocities.
there is a good agreement between the results.
Numerical buckling shapes together with axial force time curve of
With investigation on buckling shapes for S.Th tubes presented in U.Th tube under axial impact at velocity 63.41 (m/s) are shown in
Fig. 7, it is found that for velocity 30.12 (m/s), dynamic buckling with Fig. 11. All four wrinkles are initiated at a time range of 0–107 (μs). It
mild folds and for other velocities dynamic progressive buckling should be noted that wrinkles formed at both ends are main folds and
occurred. It is worth mentioning that for S.Th tube, at velocities two middle wrinkles are very mild. From 107(μs) to the end of impact
41.69 and 50.93 (m/s), progressive buckling with one fold and at process, greatest wrinkle continues to grow, whereas, other wrinkles do
velocities 58.11 and 63.41 (m/s), progressive buckling with two folds not change considerably.
occurred. Also, numerical buckling shapes together with axial force time
curve of S.Th tube under axial impact at velocity 63.41 (m/s) are

182
R. Rajabiehfard et al. Thin–Walled Structures 109 (2016) 174–184

Fig. 15. Energy absorption versus shortening for the same velocity and different thickness distribution.

depicted in Fig. 12. This figure demonstrates that both wrinkles at 5.3. Energy absorption of U.Th, S.Th and FG.Th tubes
smallest cross section initiated at a time range of 0–135 (μs). From
135 (μs) until 383 (μs), radial displacements of two formed wrinkles Numerical energy absorption results of U.Th, S.Th and FG.Th tubes
and shortening of the tube are reached to their final values, approxi- subjected to axial impact at 30.12, 41.69, 50.93, 58.11 and 63.41 (m/s)
mately. From 383 (μs) till the end of impact process, radial displace- velocities are presented in Table 2. Also, energy absorption versus
ments of middle cross section are increased. shortening for the same thickness distribution, different velocities and
Numerical buckling shapes together with axial force time curve of energy absorption versus shortening for the same velocity and different
FG.Th tube under axial impact at velocity 63.41 (m/s) are depicted in thickness distribution are depicted in Figs. 14 and 15, respectively.
Fig. 13. The greatest wrinkle is created at a time range of 0–140 (μs). According to the results of Table 2, the energy absorption for three
From 140 (μs) to 185 (μs), the created wrinkle is pulled down and from different thickness distributions, do not have significant difference.
185 (μs) until the end of impact process, the second wrinkle is created. Likewise, pursuant to the results of Table 2 and Fig. 14, it is found out

183
R. Rajabiehfard et al. Thin–Walled Structures 109 (2016) 174–184

that with increasing of velocity and constant remaining of other [2] D. Karagiozova, N. Jones, Dynamic effects on buckling and energy absorption of
cylindrical shells under axial impact, Thin-Walled Struct. 39 (2001) 583–610.
parameters, the energy absorption increases as it was predicted. [3] D. Karagiozova, N. Jones, On dynamic buckling phenomena in axially loaded
As Fig. 14 indicates, with increasing of velocity in the same elastic- plastic cylindrical shells, Int. J. Non-Linear Mech. 37 (2002) 1223–1238.
thickness distribution, the slope of the energy absorption curve [4] Y.S. Tai, M.Y. Huang, H.T. Hu, Axial compression and energy absorption
characteristics of high-strength thin-walled cylinders under impact load, Theor.
approximately remains constant. In other words, the behavior of the Appl. Fract. Mech. 53 (2010) 1–8.
energy absorption curve in the same thickness distribution and [5] D.H. Chen, K. Ushijima, Estimation of the initial peak load for circular tubes
different velocity remains constant, approximately. subjected to axial impact, Thin-Walled Struct. 49 (2011) 889–898.
[6] U. Lepik, On plastic buckling of cylindrical shells struck axially with a mass, Int. J.
Fig. 15 illustrates the behavior of the energy absorption curve in the Non-Linear Mech. 33 (1998) 235–246.
same velocity and different thickness distribution. The U.Th tube in [7] D. Karagiozova, N. Jones, Dynamic elastic-plastic buckling of circular cylindrical
comparison with S.Th and FG.Th tubes, absorbs the same energy in the shells under axial impact, Int. J. Solids Struct. 37 (2000) 2005–2034.
[8] A. Wang, W. Tian, Mechanism of buckling development and strain reversal
less shortening. This figure reveals that overall behavior of energy
occurrence in elastic-plastic cylindrical shells under axial impact, Int. J. Non-Linear
absorption versus shortening for S.Th and FG.Th tubes is approxi- Mech. 43 (2008) 722–732.
mately identical. [9] S. Salehghaffari, M. Tajdari, M. Panahi, F. Mokhtarnezhad, Attempts to improve
Generally, according to Figs. 10 and 15, it turns out that S.Th and energy absorption characteristics of circular metal tubes subjected to axial loading,
Thin-Walled Struct. 48 (2010) 379–390.
FG.Th tubes compared to S.Th tube absorb the same energy with larger [10] F. Mokhtarnezhada, S. Salehghaffaria, M. Tajdarib, Improving the crashworthiness
shortening and less peak load. So, using of S.Th and FG.Th tubes characteristics of cylindrical tubes subjected to axial compression by cutting wide
compared to U.Th tube as energy absorption system especially in grooves from their outer surface, Int. J. Crashworthiness 14 (6) (2009) 601–611.
[11] S.E. Mirmohammadsadeghi, Khalili Kh, S.Y. Ahmadi, S.J. Hosseinipour,
automotive industries, leads to the same energy absorption and less Experimental and finite element simulation investigation of axial crushing of
force transfer onto the travelers. Thus, S.Th and FG.Th tubes are better grooved thin-walled tubes. international, J. Adv. Manuf. Technol. 77 (2015) 1627–
energy absorption specimens. 1643.
[12] A. Darvizeh, M. Darvizeh, R. Ansari, A. Meshkinzar, Analytical and experimental
investigations into the controlled energy absorption characteristics of thick-walled
6. Conclusion tubes with circumfrential grooves, J. Mech. Sci. Technol. 28 (10) (2014) 4199–
4221.
[13] A. Darvizeh, M. Darvizeh, R. Ansari, A. Meshkinzar, Effect of low density, low
In this paper, the dynamic behavior of uniform thickness, stepped
strength polyurethane foam on the energy absorption characteristics of circum-
thickness and functionally graded thickness tubes under axial impact frentially grooved thick-walled circular tubes, Thin-Walled Struct. 71 (2013) 81–
were investigated experimentally and numerically. Tubes were im- 90.
[14] Z. Liu, W. Hao, J. Xie, J. Lu, R. Huang, Z. Wang, Axial-impact buckling modes and
pacted by a striking mass. Gas gun was used to perform experimental
energy absorption properties of thin-walled corrugated tubes with sinusoidal
tests; in addition, numerical results were obtained by FE simulation. patterns, THin-Walled Struct. 94 (2015) 410–423.
The effect of thickness distribution on shortening, energy absorption [15] C. Kılıçaslan, Numerical crushing analysis of aluminum foam-filled corrugated
and buckling shape of tubes were investigated. It is found that a change single- and double-circular tubes subjected to axial impact loading, Thin-Walled
Struct. 96 (2015) 82–94.
in thickness distribution of the tube is able to convert the buckling [16] A. Moradpour, M. Elyasi, S. Montazeri, Developing a new thin-walled tube
shape from buckling with mild folds to progressive buckling and vice structure and analyzing its crushing performance for aa 60601 and mild steel under
versa. Furthermore, it is found that stepped thickness tube can be an axial loading, Trans. Indian Inst. Met. (2015). http://dx.doi.org/10.1007/s12666-
015-0629-2.
approximate of the functionally graded thickness tube, which in this [17] X. Zhang, H. Zhang, Z. Wen, Axial crushing of tapered circular tubes with graded
case almost, their behavior is adopted with each other. This study thickness, Int. J. Mech. Sci. 92 (2015) 12–23.
reveals that stepped thickness and functionally graded thickness tubes [18] G. Li, F. Xu, G. Sun, Q. Li, A comparative study on thin-walled structures with
functionally graded thickness (FGT) and tapered tubes withstanding oblique impact
in comparison with uniform thickness tube, absorb the same energy loading, Int. J. Impact Eng. 77 (2015) 68–83.
with more shortening and less peak load; so, they are better energy [19] A. Alavi Nia, J.H. Hamedani, Comparative analysis of energy absorption and
absorption specimens. With comparing of experimental and numerical deformations of thin walled tubes with various section geometries, Thin-Walled
Struct. 48 (2010) 946–954.
results, it is found that there is a good agreement between them. [20] R. Gumruk, A numerical investigation of dynamic plastic buckling behavior of thin-
walled cylindrical structures with several geometries, Thin-Walled Struct. 85
References (2014) 388–397.
[21] N. Jones, Structural Impact, Second edition, Cambridge University Press,
Cambridge, 2012.
[1] A.L. Florence, J.N. Goodier, Dynamic plastic buckling of cylindrical shells in
sustained axial compressive flow, J. Appl. Mech. 35 (1968) 80–88.

184