See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/316747880

Structures under Impact Loading

Thesis · June 2005

CITATIONS READS

0 3,020

1 author:

John Kam-wing Wong

University of Cambridge
6 PUBLICATIONS   39 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Lattice Element Method View project

All content following this page was uploaded by John Kam-wing Wong on 08 May 2017.

The user has requested enhancement of the downloaded file.

 Final Year Project: Structures Under Impact Loading

Abstract

Impact loading analysis have wide applications in civil engineering including crash
barrier design , impact resistance structural design against military and terrorist attack ,
ship impact resistance design for marine structures and blonder fence design. After
911 aircraft impact incident, there is increased concern in structural response under
impact loading also.

Researches in structural analysis of impact loading are focused on the use of

non-linear finite element method (FEM) which requires large amount of computation
effort. In this report, a simpler and more intuitive beam vibration model was
suggested to assess the impact resistance of structures.

Successively more accurate models, massive beam model, effective mass model and
beam vibration model, were used to predict the time history of impact force. Beam
vibration model, which is capable to predict both local deformation and global
structural vibration of structure under impact loading, was investigated in depth. This
was done by coupling contact law model with linear multi-degree of freedom (MDOF)
dynamic model. Governing equation of beam vibration model was formulated and
numerical method was suggested to solve the equation. Two non-dimensional
parameters, mass ratio and pulse ratio governing impact behavior were identified and
investigated in parametric study.

Simply supported perspex beams under mid-span impact of rigid steel striker were
tested. Acceleration history of beam at mid-span and maximum mid span beam
displacement were measured and compared with the predictions done by beam
vibration model.

Finally, some modeling techniques using beam vibration model are discussed for
practical applications.

1
 Final Year Project: Structures Under Impact Loading

Acknowledgement

I would like to acknowledge my supervisor, Dr. R.K.L. Su, for his kind support and

guidance throughout the project. His advice on my project gave me a clear direction

in my research work. I would also like to acknowledge Mr. Shih for his effort on

preparing the electronic equipments required in experiments and the laboratory

technicians in structural laboratory of our department for their kind assistance in

preparing experimental setup.

2
 Final Year Project: Structures Under Impact Loading

Table of Content

Abstracts ..................................................................…………………………………….1
Acknowledgement ………………………………...…………………………………….2
Table of Content………………………………………………………………………….3

1 Fundamental Concepts of Impact ……….………………………………………4 - 8

2 Beam Vibration Model………………………..………………………………...9 -26

3 Parametric Study………………………………………………………………27 - 37

4 Experimental Investigation …………………...………………………………38 - 47

5 Application of Beam Vibration Model………..………………………………49 - 51

6 Conclusion & Suggestion……………………..………………………………52 – 53

Reference ………………………………………..…………………………………...54

Appendix A Numerical Solution for Impact Solution

Appendix B Tables & Graphs for Parametric Study
Appendix C Photos of Experimental Setup
Appendix D Graphs for Beam Impact Test Experimental Result
Appendix E Graphs for Tensile Tests
Appendix F Graphs for Indentation Tests

3
 Final Year Project: Structures Under Impact Loading

CHAPTER ONE

Fundamental Concepts for Impact

When 2 bodies hit with each other, the surface of two bodies come together with some
relative velocity at an initial instant termed incidence. After incidence there would be
interference or interpenetration of the bodies of the bodies to build up interface
pressure that arises in a small area of contact between two bodies. This small area is
called impact zone in this report. At each instant during the contact period, the
pressure in the contact area results in local deformation and consequent indentation.

At each instant during impact the contact pressure has a resultant force acting in
opposite directions on the two colliding bodies which is called impact force. Initially,
the impact force increases with increasing indentation and it reduces the speed at
which the bodies are approaching each other. At some instant during the impact the
work done by the contact force is sufficient to bring the relative velocity of two
bodies to zero. Subsequently, the energy stored during compression drives the two
bodies apart until they separate with some relative velocity. For impact between solid
bodies, the impact force that acts during collision is a result of the local indentations
that are required for the surfaces of two bodies to conform in the contact area.

The local indentations that arise during impact vary according to the incident relative
velocity at the point of initial contact and the hardness of the colliding bodies. Low
speed collision result in contact pressures that cause small indentations and significant
only in a small region only. For high speed impact, for example bullet hitting metal
plates, there are large deformations near the contact area near the contact area which
result from plastic flow. In most of applications in civil or structural engineering,
impacts are classified as low speed impact. Therefore, low speed impact was
investigated in this project

The collision is modeled as elastic so impact forces can be assumed to be conservative

or reversible (i.e. loading and unloading curve are identical). Therefore, impact force

4
 Final Year Project: Structures Under Impact Loading

is a function solely of the local indentation of the interacting bodies. Friction, which
acts tangentially at the impact zone, is assumed to be negligible. Followed by these
assumptions, the relationship between impact force and local indentation can be
modeled by different contact laws.

1.1 Types of impact

Impact can be classified into 3 types: particle impact, rigid body impact and
transverse impact on flexible bodies.

Particle Impact

Particle impact is an analytical approximation that considers a normal component of

impact force impulse only. The source of force is unspecified, but presumably it is
strong and the force has a very short duration, so the duration of impact is negligible
small instant of time. If only the kinematics of impact is concerned and the global
structural vibration is negligibly small, particle impact can provide a very simple
solution by considering conservation of momentum and law of restitutions.

Rigid Body Impact

It occurs between compact bodies where the contact area remains small in comparison
with all section dimensions. Stress generated in the contact area decrease rapidly with
increase radial distance from the contact region. The indentation is small and confined
in a small impact zone. Apart from the kinematics of impact, the time history of the
impact force can be obtained by various contact laws. Massive beam model and
effective mass model can be used to model rigid body impact.

Transverse impact on flexible bodies

It occurs if one of the bodies suffers bending as a result of impact force. The bending
reduces the impact force as the ‘stiffness’ of impact zone reduces. Subsequence
impacts may occur because of the vibration induced. Structural vibration is a source
of energy dissipation during collision. Beam vibration model is used to model this
type of impact.

1.2 Hertz Law of Contact

Contact laws are the basis in the theoretical study of impact as it relates the impact
force with the local indentation. There are many contact laws developed for impact
with different characteristic. Hertz contract law is the most common one among them.
It provides a very good approximation for collision between hard compact bodies
where contact region remains small.

Here are the major assumptions in Hertz contact law:

1. The surface of the impactor at impact zone is spherical

2. Both the impactor and the beam remain elastic throughout the impact
3. Materials at impact zone behave identically during approach and restitution.

5
 Final Year Project: Structures Under Impact Loading

Consider a rigid mass strike on a beam which is assumed to be immovable in normal

direction. It can be proved analytically that the impact force F is proportional to the
3/2 power of indentation a at any instant of impact time t

F (t ) = Ka 3 / 2 (t ) ………………….(1-1)

F(t)
a

K is called contact constant which is governed by the elastic and geometry properties
of two bodies. If the striker has negligible deformation compared with the beam, the
radius of striker, the Young Modulus of the beam and the thickness of beam are major
factor determines the contact constant K

1.3 Massive Beam Impact

The beam impact problem can be greatly simplified if the mass of the beam is
sufficiently greater than the mass of striker. The velocity of beam as a whole is
negligible compared with the local deformation at impact zone.

vo ms
z– Initial separation of center of
At t =0 masses of both striker and
z
beam

ms
z-a(t) During Impact
Impact Force F(t)

Consider a rigid spherical striker with mass ms striking the massive beam at velocity
vo. The dynamic equilibrium of striker yields the following equation

F − m s a&& = 0 ………………….(1-2)

Using Hertzian contact law F = Ka 3 / 2 , the above equation can be written as

Ka 3 / 2 − m s a&& = 0 ………………….(1-3)

6
 Final Year Project: Structures Under Impact Loading

This second order ordinary differential equation can be solved for maximum
indentation, maximum impact force and corresponding time.

Maximum indentation 2/5

 5m v 2 
a∞ =  s o 
 4K 
 
Maximum impact force 3/5
 5m s v o 2 K 2 / 3 
F∞ =  

 4 
Corresponding time of maximum impact 2/5
 5m s 
force t ∞ = 1.47 1/ 2


 4 Kvo 
The impact force history can be  πt 
2/5

approximated by F = F∞ sin  
 2t ∞ 

The massive beam model is regarded as the upper bound solution of impact problem.
In this model, the kinetic energy of striker is converted into the elastic energy at the
impact zone. The kinetic energy of beam, energy lost due to inelastic deformation at
impact zone and the structural vibration induced during impact are neglected.

1.4 Effective Mass Impact

The effective mass model modifies the massive beam model by introducing one more
parameter - mass of beam mb. During impact, the impact force accelerates the beam
which is initially at rest. By considering relative velocity between the striker and the
beam, the beam can be modeled as if it is immovable. This is done by reducing the
mass of striker in the massive beam model .By conservation of momentum, the mass
ms me  α 
of the striker ms in the massive beam mode is replaced by or  ms if
m s + me 1+ α 
a mass ratio α is defined as the ratio between effective mass of beam me to the mass
of striker ms.

From the governing equation of massive beam impact:

 α 
F − m s a&& = 0 ………………….(1-4)
1+ α 

The following important result can be obtained:

am  α 
2/5

Maximum indentation = 
a∞  1 + α 
Fm  α 
3/5

Maximum impact force = 

F∞  1 + α 
2/5
tm  α 
Corresponding time of maximum impact force =  
t∞ 1+ α 
 

7
 Final Year Project: Structures Under Impact Loading

The following show parametric study of effective mass model:

Time History of Impact Force

18000 Massive Beam Solution

16000 Mass Ratio = 5

14000 Mass Ratio = 1

Mass Ratio = 0.2
Impact Force(N)

12000

10000

8000

6000

4000

2000

0
0 0.1 0.2 0.3 0.4 0.5 0.6
Time (ms)

Effect of inertia in Effective Mass Model

1
Fm
0.9
F∞
0.8
0.7
Output parameter

am
0.6 a∞
0.5
0.4
0.3
0.2
0.1
0
0 2 4 6 8 10
Mass ratio mb/ms

8
 Final Year Project: Structures Under Impact Loading

CHAPTER TWO

Beam Vibration Model

Beam vibration model is a more accurate model which describes the mechanics of
impact. In addition to the inertia of beam, the model also considers transverse
deformation of beam (flexural stiffness of beam) during impact.

The vibration of beam induced by impact is modeled as multi-degree of freedom

(MDOF) undamped dynamic system. This is regarded as global behavior as all beam
elements contribute in the vibration in this dynamic system. The global behavior of
beam is governed by the flexural stiffness of beam and inertia properties of beam.

The local indentation which is the basis of impact theory is modeled by Hertz Contact
law. This is regarded as the local behavior as only beam elements at impact zone are
affected.

2.1 Global behavior of beam

In beam vibration model, multi degree of freedom (MDOF) undamped dynamic

system is used to model the behavior of beam under impact loading. The duration of
impact force is very small compared with periods of vibration modes excited. An
undamped dynamic system is used because damping force has negligible effect for
such a short duration impact force. This will be illustrated by considering a SDOF
damped system shown below. The damping is assumed to be viscous (e.g. damping
force is proportional to the velocity) which is a good approximation for most of
structural system.

9
 Final Year Project: Structures Under Impact Loading

External Force:
F (t )

Inertia Force:
m&y&(t )

Spring Force:
ky (t ) Damping Force:
cy& (t )

The governing equation of a damped dynamics system:

m&y& + cy& + ky = F (t ) ………………….(2-1)

where m : mass of body

c : damping constant
k : spring constant

During impact, the magnitude of impact force is very large but it lasts for a very short
period of time. The mass accelerates to reach certain velocity in a very short period of
time and hence the acceleration, which is the time derivative of velocity, is much
larger. The damping force, cy& , is negligible compared with the inertia force m&y& .

Comparsion between damped dynamic model and undamped dynamic model

5 . 0 0E - 0 3

4 . 0 0E - 0 3

3 . 0 0E - 0 3

2 . 0 0E - 0 3
Displacement

1 . 0 0E - 0 3

0 . 0 0E + 0 0

0 2 0 0 4 0 0 6 0 0 8 0 0 1 0 00 1 2 0 0

- 1 . 0 0E - 0 3

- 2 . 0 0E - 0 3

- 3 . 0 0E - 0 3

- 4 . 0 0E - 0 3

Damped (2.5%)
- 5 . 0 0E - 0 3

Time Undamped

As shown above, the effect of damping became significant after few cycles of

10
 Final Year Project: Structures Under Impact Loading

vibration. For higher vibration modes, damping is more significant as it goes through
more vibration cycles during impact. However, such high modes of vibration are very
unlikely to be excited during impact.

2.2 Multi-Degree of freedom system

Theoretically, infinity many vibration modes can be induced by excitation force. In

most of structural dynamics analysis in common engineering problems, single
freedom of freedom (SDOF) dynamic models are often used to simplify the analysis.
In this model, it is assumed that the vibrations induced are dominated by first
vibration mode. This gives a reasonable accurate approximation if the period of
excitation is the same order of magnitude of fundamental period of structure.

In most of impacts, the natural period of beam can be a hundred times longer than the
impact duration. More energy is transferred to higher mode vibrations during impact
because the periods of higher mode vibration are close to the duration of impact force.
A multi-degree of freedom system is necessary to model the dynamic behavior of
beam accurately.

From table below, we can observe that high vibration mode induces low displacement,
but high shear force. Also, high vibration modes induce high acceleration rate but low
displacement.

Static mode for mid-span Free Vibration mode for pinned end beam
loading 1st mode 3rd mode

0.11

1 0.99

0.82 0.82
1

1.28 3.84
1

Comparison of pin ended beam modes with equal potential energy

The pulse ratio β give a good indication on the number of higher vibration modes that

11
 Final Year Project: Structures Under Impact Loading

are significant to the beam vibration model. The graphs below show the model
participation in systems with different pulse ratio.
Comparsion of beam displacement by including more vibration modes (b=0.0035)

0.005

29 modes
0.004 5 modes
3 modes
Displacement (m)

0.003 1 mode

0.002

0.001

0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Time (ms)

Comparsion of beam displacement by including more vibration modes

(b=0.05)

0.004 29 modes
0.0035 5 modes
Mid span displacem ent (m)

3 modes
0.003
1 modes
0.0025

0.002
0.0015
0.001
0.0005
0
0 0.2 0.4 0.6 0.8 1
Time (ms)

12
 Final Year Project: Structures Under Impact Loading

Comparsion of beam displacement by including more vibration modes

(b=0.128)

0.003
Mid-span displacement (m)

0.0025

0.002

0.0015

0.001 29 modes
5 modes
0.0005
3 modes
0 1 modes
0 0.2 0.4 0.6 0.8 1
time (ms)

13
 Final Year Project: Structures Under Impact Loading

2.3 Free Vibration Analysis

Consider a vibrating beam of length L

x ∂2 y
ρA∆x
0 L ∂x 2
∆x

y
Q
Q+∆Q
M M+∆M

Consider a beam element of length ∆x, the inertia force is expressed as

∂2 y
F∆x = ρA∆x ………………….(2-2)
∂x 2

For moment equilibrium of a beam element

(2Q + ∆Q ) ∆x = (M + ∆M ) − M ………………….(2-3)
2

Rearranging,

∂M ∂3 y
Q= = − EI 3 ………………….(2-4)
∂x ∂x

Differentiate with respect of x once

∂Q ∂4 y
− EI 4 ………………….(2-5)
∂x ∂x

For force equilibrium of the beam element

∂2 y
(Q + ∆Q ) − Q = ρA∆x 2 ………………….(2-6)
∂t

Rearranging,

∂Q ∂2 y
= ρA 2 ………………….(2-7)
∂x ∂t

14
 Final Year Project: Structures Under Impact Loading

From equation (2-5) and (2-7),

∂4 y ∂2 y
EI + ρA =0 ………………….(2-8)
∂x 4 ∂t 2

∂4 y
Note that from equation (2-8), EI is the spring force from the flexural stiffness of
∂x 4
∂2 y
beam and ρA is the inertia force. Because there is no external force acting on the
∂t 2
beam element in free vibration analysis, the spring force and the inertia force always
balance with each other. The differential equation can be solved using separation of
variable technique shown as follow.

To solve the different equation (2-8), the beam displacement at any position x and
time t can be express into

∞ ∞
y ( x, t ) = ∑ y i = ∑ φi ( x )Yi (t ) ………………….(2-9)
i =0 i =0

where φi(x) is the i th vibration mode shape (modal shape) which is the eigen function
of the governing differential equation of free vibration. Yi(t) gives the magnitude of
the vibration of i th vibration mode at any time t

Substitute (2-9) into (2-8)

∂ 4φ ( x ) ∂ 2Yi (t )
EIY (t ) + ρAX ( x ) =0 ………………….(2-10)
∂x 4 ∂t 2

Rearranging,

EI φi'''' Y&&
= − = ω i2 ………………….(2-11)
ρA X i Yi

where ϖi is an arbitrary constant. From (2-11), the differential equation (2-8) can be
separated into 2 ordinary different equations below.

X i + ζ 2ω 2 X i = 0
' '''
………………….(2-12)
Y&& + ω 2Y = 0
i i
………………….(2-13)
EI
where ζ =
2

ρA
Solving the above equation gives
 ω   ω   ω   ωi 
φi (x ) = Ai sin  i  + Bi cos i  +C i sinh  i  + Di cosh 
  ζ 
 ….(2-14)
 ζ   ζ   ζ   
Yi (t ) = Ei sin ω i t + Fi cos ωi t ….(2-15)

15
 Final Year Project: Structures Under Impact Loading

The constants Ai , Bi , Ci and Di are determined from the beam boundary conditions.
Notice that they cannot be solved explicitly, and a value have to be assumed to one of
the constants. The constant Ei and Fi are determined from initial condition. For i=0, it
refer to a rigid body motion of the beam (no relative velocity between beam elements).
The eigen value of the governing equation, ϖi is the period of ith vibration which is
the time required to complete one cycle of vibration. The vibration mode shape and
period govern the dynamic behavior of the beam.

Now, we consider a pin-ended beam. The vibration mode can be determined by the
following boundaries

At x=0 and x=L,

∂ 2 yi
yi = 0 and =0 for i=1,2,…
∂x 2

The constants Ai , Bi , Ci and Di are determined. The period ϖi can also be determined

∞
 iπx 
[Ei sin ω i t + Fi cos ω i t ]
2
y=∑ sin  ……………….(2-16)
i =1 L  L 

i 2π 2 EI
where ω i =
2

L2 ρA
By applying different boundaries conditions, different vibration modes and their
corresponding vibration period can be found.
θ 2 EI
Value of υ from ω i = 2
2

L ρA
Pin-ended beam Cantilever beam Fixed end

3.142 4.730
1.875

7.853
6.283 4.694

9.425 10.996
7.855

12.566 10.996 14.137

16
 Final Year Project: Structures Under Impact Loading

2.4 Force vibration of a beam

The beam impact problem can be regarded as forced vibration of beam with impact
force as an external force. The governing equation of the forced beam vibration is
formulated below :

w(x,t)
x
w(x,t)
0 L
∆x

y
Q
Q+∆Q
∂2 y
M ρA∆x M+∆M
∂x 2

For moment equilibrium of a beam element

(2Q + ∆Q ) ∆x = (M + ∆M ) − M ………………….(2-17)
2

Rearranging,

∂M ∂3 y
Q= = − EI 3 ………………….(2-18)
∂x ∂x

Differentiate with respect of x once

∂Q ∂4 y
− EI 4 ………………….(2-19)
∂x ∂x

For force equilibrium of the beam element

(Q + ∆Q ) − Q + w(x, t )∆x = ρA∆x ∂

2
y
………………….(2-20)
∂t 2

Rearranging,

∂Q ∂2 y
= ρA 2 − w( x, t ) ………………….(2-21)
∂x ∂t

Substitute (2-19) into (2-21)

∂4 y ∂2 y
EI + ρA = w( x, t ) ………………….(2-22)
∂x 4 ∂t 2

17
 Final Year Project: Structures Under Impact Loading

Compare with the governing equation in free vibration of beam, there is an additional
term w(x,t) which is the external force per unit length. To solve the governing
equation,

∞ ∞
y ( x, t ) = ∑ y i = ∑ φi ( x )Yi (t ) ………………….(2-23)
i =0 i =0

where φi (x ) is the modal shape from the free vibration analysis discussed before and
Yi(t) is the modal participation which give the magnitude of vibration.

∞ ∞
d2  d 2φ i 
∑ ρAφi ( x )Yt (t ) + ∑ 2
&&
 EI Yi (t ) = w(x, t ) ………………….(2-24)
i =1 i =1 dx  dx 2 

Multiply each term by φ n ( x ) and integrating gives

∞ ∞
&& (t ) L ρAφ ( x )φ ( x )dx + Y (t ) L φ (x ) d
2
 d 2φ i 
∑ ∑ dx = ∫ φ n (x )w( x, t )dx
L

i =1
Yt ∫0 i n i =1
i ∫0 n dx 2  EI

2 
dx  0

…………….(2-25)

By the orthogonality of mode shapes, the integration of φi ( x )φ n ( x ) vanishes except

for i=n. therefore,

d 2  d 2φ n 
Yt (t )∫ ρAφ n ( x )dx + Yi (t )∫ φ n (x ) 2  EI ∫0 φ n (x )w(x, t )dx
L L L
&& =
2
dx
0 0 dx  dx 2 
…………….(2-26)
d  d φn 
2 2
The term ∫ φ n ( x ) 2  EI dx can be expressed as ω n2 ∫ ρAφ n (x )dx by
L L 2
2 
0 dx  dx  0

multiplying equation (2-8) in free body analysis with φ n (x ) and carrying out
integration along the beam.

By introducing the following three terms:

ρAφ n 2 ( x )dx
L
Generalized Mass Mn: ∫ 0

d2  d 2φ n 
φ n (x )
L
Generalized Stiffness Kn: ∫  EI 2 
dx = ω n M n
0 dx 2  dx 

∫ φ (x )w(x )dx
Generalized Loading Wn: L
n
0

18
 Final Year Project: Structures Under Impact Loading

The equation can be written as

M n Y&&n (t ) + K n Yn (t ) = Wn ………………….(2-27)
2
 ∂ 2φ 
EI ∫  2  dx
0 ∂x
where
ω2 =   period of nth vibration mode
L
ρA∫ φi dx 2
0

For every vibration mode, equation (2-26) can be established and solved for
corresponding modal participation (magnitude of vibration mode).

Compare equation (2) with equation for SDOF dynamic model:

m&y& + ky = F (t ) ………………….(2-28)

The MDOF dynamic model is decoupled into a number of SDOF systems. The modal
participation in each SDOF system can be solved by Duhamel equation which
provides analytical solution for any type of excitation force.

The Duhamel solution of equation (2-26) is

Yn (t ) = ∫ W (τ )sin ω (t − τ )dτ
1 t
………………….(2-29)
M nω n
n i
0

The vibration of beam y(x,t) can be found by superposition of all vibration modes.

∞
φi (x )
y ( x, t ) = ∑  L w(τ )φ (x )dx  sin ω (t − τ )dτ
t

i =1 ρAω i ∫
∫0 ∫0
φ 2 ( x )dx 
i  i …………….(2-30)
i

If the excitation force can be regarded as a point load F(t) acting at x=xo. Equation
(2-30) becomes

∞
φi ( x )φ ( xo )
y ( x, t ) = ∑ ∫ F (τ )sin ω (t − τ )dτ
t

(x )dx ………………….(2-31)
i
ρAω i ∫ φ 2
0
i =1

For a mid-span impact of a pinned end beam, the vibration modes are

 iπx 
φi (x ) =
2
sin   for i=1,3,…
L  L 
Note that only odd vibration modes will be excited for mid span impact.

The following terms can be found

L
∫ φi2 dx = 1 ………………….(2-32)
0

19
 Final Year Project: Structures Under Impact Loading

 ∂ 2φ i (x ) 
2 ………………….(2-33)
 iπ 
4
L
∫ 0  ∂x 2  dx =  L 




so

2
 ∂ 2φ 
EI ∫  2  dx
0 ∂x
  i 4π 4 EI ………………….(2-34)
ω2 = =
L
ρA∫ φ i2 dx L4 ρA
0

and the beam displacement

∞
iπx t
y ( x, t ) = F (τ )sin ω i (t − τ )dτ
2 1
mb
∑ω i =1
sin
L ∫0
………………….(2-35)
i

In particularly at mid-span,

∞
L  2
∑ ω ∫ F (τ )sin ω (t − τ )dτ
1 t
y , t  = i ………………….(2-36)
 2  mb i =1 i
0

2.5 Impact between striker and beam

Consider the impact of a rigid striker, of mass ms and initial velocity vo with a uniform
single-span beam initially at rest. The beam is pin ended with a span length L. The
impact occurs at impact point x=xo, as shown in figure below. To simply the problem,
a mid-span impact (i.e xo=L/2) is considered so unsymmetrical vibration mode will
not be excited by the impact. Assume the impact zone obeys Hertz contact law
with F (t ) = Ka 3 / 2 (t ) . It is further assumed that the global beam behavior can be
described by a vibration mode as discussed in the previous section.

ms vo
x
mb

The key step in formulating the governing equation is to couple the effect of contact
law with dynamics system described in pervious section. This is done by the
compatibility of displacement between the striker and the beam.

20
 Final Year Project: Structures Under Impact Loading

y(x,t)
ys(t)
a

Compatibility equation at impact zone:

a(t ) = y s (t ) − y ( xo , t ) ………………….(2-37)

where a (t ) local indentation at impact zone (Local behavior of beam)

y s (t ) displacement of striker at time t
y (xo , t ) displacement of beam at point of impact xo at time t (Global
behavior of beam

There are three terms in the compatibility equation. All of them are displacement. To
solve this equation, the compatibility equation has to be re-written as a function of
impact force F(t) which is in common in all three terms.

For the first term, consider the Hertz contact law

F (t ) = Ka 3 / 2 (t )

Then,

 F (t ) 
2/3

a(t ) =   ………………….(2-38)
 K 

To express second term and third term in compatibility equation in terms of impact
force F(t), consider the free body of strikers and beam separately.

y(x,t) ms &y&s (t )
F(t)
F(t)

Inertia force +
Beam spring force
Free body of striker
Free body of beam
Consider the dynamics equilibrium of striker:

F (t ) − m s &y&s (t ) = 0 ………………….(2-39)

21
 Final Year Project: Structures Under Impact Loading

Rearranging,
F (t )
&y&s (t ) = ………………….(2-40)
ms

Carry out integration with respect to time twice and apply boundary conditions
ys(0)=0 and y& s (0) = vo ,

1 t τ
y s (t ) = v o t − ∫∫ F (t ) d tdτ ………………….(2-41)
ms 0 0

Equation (2-36) deduced before describes the displacement of beam under impact
loading. Putting equation (2-36), (2-38) and (2-41) into the compatibility equation, the
governing equation of beam impact is formulated as below:

2/ 3 ∞
 F (t )  1 t τ 2 1
∑ ω ∫ F (τ ) sinω (t − τ )dτ
t
  = vo t − ∫∫ F (t )d tdτ − i
 K  ms 0 0 mb i =1,3 i
0

………………….(2-42)

The governing equation is non-linear as the Hertz contact law is non-linear. So, the
equation (2-42) cannot be solved analytically. Numerical techniques have to be
employed to obtain the force history of impact force. The integrations in above
equation are replaced by summation over a finite number of time interval ∆t. Impact
force are assumed to vary linearly within time intervals. By this approximation,
impact force at different time interval can be solved. The details of numerical method
employed in this project is present in Appendix A.

Beam displacement can be calculated by back substitute the time history of impact
force into equation (2-35). Shear force, bending moment at mid-span can be found by
the following equation

 L  2ω L ∞ t
Mid-span moment : M  , t  = 12 ∑ ∫ F (τ ) sin ωi (t − τ )dτ
 2  π i =1,3 0
Mid-span shear :  L  2ω ∞ iπ t
Q , t  = 21 ∑ i sin ∫ F (τ ) sin ωi (t − τ )dτ
 2  π i =1,3 2 0

Kinetic energy Ui and potential energy Vi corresponding to vibration mode i can be

found as follow:

1  ∂yi 
2 2
U i (t ) = ∫
L 1  t 
Mid-span moment : ρA  dx =  ∫ F (τ ) cosωi (t − τ )dτ 
0 2  ∂t  mb  0 
2
1  ∂ 2 yi  2
Q(t ) = ∫
1  t
Mid-span shear : EI  2  dx =  ∫0 F (τ ) sin ωi (t − τ )dτ 
2  ∂x  mb  

22
 Final Year Project: Structures Under Impact Loading

The striker velocity vs after impact can be computed also

∫ F (τ )dτ
t

Striker velocity vs v s = vo − 0

ms

2.6 Impact criteria

In effective mass impact, the beam and the striker do not have second impact. In beam
vibration model, there may be subsequence impacts after the first impact due to the
rebound. The time and the time history of subsequence impact can be determined by
considering impact criteria.

Impact criteria: at any time t

y s (t ) ≥ y (x o , t ) ……Impact occur
y s (t ) < y (x o , t ) ……No impact

Consider the compatibility equation,

a(t ) = y s (t ) − y ( xo , t )

By Hertz contact law,

 F (t ) 
2/3

  = y s (t ) − y ( xo , t ) ………………….(2-43)
 K 

Mathematically, the right hand size of equation (2-43) must be non-negative for a real
solution of F(t).

A negative a(t) means that the striker and the beam has lost contact. Contact force
cannot exist between the striker and the beam. Impact force F(t) , as a contact force
between the striker and the beam, should vanish once the impact criteria cannot be
fulfilled. The governing equation of beam impact (2-42) is no longer valid.

From the table below, the impact criteria can be employed for both positive and
negative values of beam displacement y(xo,t) and striker displacement ys(t).

23
 Final Year Project: Structures Under Impact Loading

y s (t ) > y (x o , t ) , impact occur

x For
y(x,t) y s (t ), y (x o , t ) > 0
ys(t)
a
y

For
y s (t ), y (x o , t ) < 0
a
y(x,t)
ys(t)
x

y
For
y(x,t) y s (t ) > 0,
a x
ys(t) y(xo , t ) < 0

y s (t ) < y (x o , t ) , No impact
For
x y s (t ), y (x o , t ) > 0
ys(t) y(x,t)

For
y s (t ), y (x o , t ) > 0

ys(t) y(x,t)
x

For
y s (t ), y (x o , t ) > 0
ys(t) y(x,t)
x

24
 Final Year Project: Structures Under Impact Loading

If the system satisfies the impact criteria, the beam and the striker are coupled by
compatibility equation (2-37) and treated as a single system. Analysis mentioned in
pervious section can be adopted. On the other hand, if the system does not satisfy the
impact criteria, the beam and the striker are decoupled from compatibility equation
(2-37) and treated them as separated systems. Free body of the beam and the striker
are considered separately.

If acceleration due to gravity can be ignored, the striker will continue to travel at
velocity vs (velocity of striker just finished the impact) .The beam will undergo free
vibration with initial conditions at the end of impact. Subsequence impact occurs if
the impact criteria are satisfied again.

An impact cycle is formed - the beam contacts with the striker, impact occur, the
beam is lost contact with the striker and then contacts the striker again. This cycle will
continue indefinitely if there is no mechanism to dissipate energy away in the system.
In reality, there is energy dissipation the beam and the striker during impact and
during free vibration of beam. The kinetic energy from the striker

Full cycle of impact

The striker contact with
ms vo
the beam. Impact criteria
is satisfied. First impact
mb start with vs=vo, vb=0 and
so
vs(t)=vb(xo, t).

In approach state
x vs(t) > vb(x0,t).
vs(t)
Impact force is increasing
y and indentation is
vb(xo,t) increasing.

The impact force reach

maximum and so the
x
indentation,
vs(t)
vs(t)=vb(xo, t)
y vs(xo,t) Restitution starts

At the end of first impact,

vs(t) vs(t) < vb(xo, t)
x

Notice that the velocity of

beam element at
y mid-span can be either
positive or negative.
vb(xo,t)

25
 Final Year Project: Structures Under Impact Loading

The system does not

satisfy Impact criteria.
vs(t) The beam undergoes
x
free vibration and striker
assumed to travel at
constant velocity. At
vb(xo,t) some instant, the beam
y
starts to “chase up” the
striker. That is
vs(t) < vb(xo, t)
The beam satisfies the
impact criteria again.
vs(t)
vs(t) < vb(xo, t).
x The beam return to the
vb(xo,t) pervious condition
y

vo After few impacts , the

ms beam and the striker
come to rest.
mb

Initial Conditions

Impact criteria
y s (t i ) ≥ y (x o , t i )?

Solve beam Treat beam and

vibration equation striker separately

y s (t i ) ≥ y ( xo , t i )

Proceed to next
Impact force
Termination? time interval
History F(t) i→i+1

Flow chart summarizing the methodology of beam impact problem analysis

26
 Final Year Project: Structures Under Impact Loading

CHAPTER THREE

Parametric Study

3.1 Dimension Analysis

The impact force F at any instant is assumed to be governed by the following

parameters:

Parameter Dimensions
[ M ][ L]
1. Impact force , F
[T ] 2

2. Mass of striker , ms [M]

3. Effective mass of beam, me [M]

[ L]
4. Velocity of approach, vo
[T ]
[M ]
5. Contact Constant ,K
[T ]2 [ L]1 / 2

6. Natural period of beam , T1 [T]

where [M] is mass, [L] is length and [T] is time

i.e Fm = f1 ( ms, me, vo, K, T1)

27
 Final Year Project: Structures Under Impact Loading

Three non-dimensional parameters can be figured out by dimension analysis, they are

F
Non-dimensional impact force F’ 3/5
ms v 06 / 5 K 2 / 5
me
Mass ratio α
ms
2/5
ms
Pulse ratio β
T1 K 2 / 5 v 10 / 5

The non-dimensional impact force F’ is dependent to mass ratio α and the pulse ratio
β. Mass ratio α indicates the contribution of inertia in resisting the impact which have
T K 2 / 5 v1 / 5
presented in parameter study of effective mass model. Pulse ratio β , 1 2 / 5 0 , can
ms
be rewritten as a more meaningful form by express it in term of t ∞ , where t ∞ is the
time to maximum impact force in massive beam model.

2/5
 5m s  m s2 / 5
Recall that t ∞ = 1.47  ∝ 2 / 5 1/ 5
1/ 2  K vo
 4 Kvo 

T1
Therefore b can also be defined as . The physical significance of pulse ratio β is
2t ∞
now apparent. It is the ratio of impact duration to natural period of the beam. It
indicates how much vibration and number of vibration mode excited by impact.

3.2 Analogy of beam vibration model

Impact zone is
Local indentation is modeled as a
modeled by Shortening of upper spring:
weightless mass
a non-linear spring ms vo a(t)=ys(t)- y(xo,t)

K ys(t)
mb
y(xo,t)
Beam vibration Inertia of beam is
is modeled by k1 modeled as a
a linear spring effective mass

The mechanism of beam vibration model can be explained by above analogy. The
MDOF dynamics system is simplified to a SODF dynamics system for ease of
illustration. Contribution of higher model vibration can also be understood by

28
 Final Year Project: Structures Under Impact Loading

employing corresponding parameters in this analogy. This analogy will be used to

demonstrate how different parameters alter the impact behavior.

The beam is modeled as a concentrated mass with effective mass me (for pin-ended
beam, me= 1/2mb). A weightless mass is supported on a non-linear spring. This
non-linear spring is supported on the beam effective mass also. The weightless mass
represents top fiber of the beam at impact zone. The impact zone is represented by the
system of weightless mass and the non-linear spring. If the system satisfy Hertz
contact law, the displacement of weightless mass is proportional to 3/2 power to the
shortening of spring. The flexural stiffness of beam is modeled as a linear spring with
stiffness of first mode vibration k1.

Two spring systems can be identified, upper spring system comprised of weightless
spring and lower spring system comprised of effective beam mass and a linear spring.
The upper spring system models the local behavior of beam and the lower spring
system models the global behavior of beam. The impact mechanics is represented by
coupling these two spring systems.
Inertia
force
Impact force m s &y&e (t )
Ka 3/ 2
(t )
Impact force
Ka 3 / 2 (t )
Inertia force
Spring force me &y&(xo , t )
k1 y (xo , t )

Free body of beam Free body of striker

By constructing free bodies of beam and striker, many similarities can be found
between beam vibration model and two spring analogy. Also, this analogy has many
similarity of a 2 degree of freedom lumped mass model except there is a non-linear
spring.

When this analogy is used, we should bear in mind that the non-linear spring is much
stiffer compare with the linear spring in most of beam encountered in engineering
practice. For a 30mm steel pad, the contact constant is 200kN/mm3/2 which is much
greater than the flexural stiffness of common structural members which are only a few
kN/mm. In most circumstance, the maximum beam displacement occurs at free
vibration stage.

29
 Final Year Project: Structures Under Impact Loading

The following figures summarize the behavior of 2 spring system in an impact.

vs may be
or
vo vs may be
vs or

vb vb

vb = 0

1. Impact starts. 2. During impact, the 3. Once the striker lost 4. When all kinetic
The striker striker compresses contact with the energy in beam
and the the non-linear weightless mass, the mass converts to
non-linear spring. The upper non-linear spring potential energy in
spring form spring system then return to its original linear spring, beam
new upper transmit vibration to length. The beam mass attain its
spring lower spring system continues to vibrate. maximum
system. displacement.

3.3 Parametric Study of beam vibration model

The following investigate how the following parameters influence the local behavior
and global behavior of beam under impact : the contact constant K, effective beam
mass mb and flexural stiffness k (which is approximated by stiffness 1st vibration
mode stiffness k1) . This is done by varying one of the parameters keep remaining
parameters constant. The difference is explained by the analogy of 2 spring system.

30
 Final Year Project: Structures Under Impact Loading

A. Varying contact constant K

Maximum deflection of
beam at free vibration

ms vo

me

k1

Soft material at impact zone Hard material at impact zone

Before Impact (K is small) (K is large)

Flexural Stiffness and mass of beam are kept constant

For a system with soft material at impact zone, it is obvious that the local indentation
is greater. The beam displacement is smaller because the resistance from inertial force
and spring force is greater than the stiffness at impact zone ( the stiffness of non-linear
spring at the top).

The overall mobility of weightless mass (the impact zone) is determined by the
combined stiffness of non-linear spring, inertia of effective beam mass and the linear
spring below. During impact, the striker has greater displacement as the soft
non-linear spring reduce the overall mobility of weightless mass.

If the contact constant is reduced, the difference in vibration periods between upper
spring system and lower spring system is reduced. The 2 spring system can transfer
striker’s kinetic energy to lower spring system more efficiently. The lower system
receive more energy during impact so the maximum displacement in free vibration
stage Therefore, the system with hard material at impact zone experience greater
beam displacement.

31
 Final Year Project: Structures Under Impact Loading

B. Varying beam mass mb

***Appendix B contains all the graphs related to parametric study***

Maximum deflection of
beam at free vibration
ms vo

mb

k1

Light beam Massive beam

Before Impact (me is small) (me is large)

Contact constant at impact zone and flexural stiffness are kept constant

For a system with smaller beam mass, the inertial force is smaller. The spring force is,
however, greater in this case. The contribution of lower spring force is small
compared with inertia force which will be discussed later. For the same impact, the
smaller beam mass gives greater beam displacement.

The upper spring force has to balance the inertial force and spring force. Smaller
inertia force means smaller upper spring force and hence smaller local indentation.

During impact, the striker has greater displacement as smaller beam mass reduces the
overall mobility of weightless mass. Following the argument presented in pervious
comparison, for light beam case, maximum beam displacement in free vibration stage
is smaller due to small overall mobility of weightless mass.

32
 Final Year Project: Structures Under Impact Loading

C. Varying flexural stiffness of beam k

As the mass of beam is fixed, increasing the flexural stiffness is equivalent to

increasing the first mode circular frequency ϖ1 according to this fundamental
formula:

k1 = meω12 ………………….(?)

The flexural stiffness k is approximated by first vibration mode stiffness k1

***Appendix B contains all the graphs related to parametric study***

Maximum deflection of
beam at free vibration
ms vo

mb

k1

Flexible beam Stiff beam

Before Impact (k1 is small) (k1 is large)

Contact constant at impact zone and mass of beam are kept constant

For a system with smaller flexural stiffness, the spring force is smaller. Obviously, it
gives greater beam displacement.

The upper spring force has to balance the inertial force and spring force. Smaller
spring force means smaller upper spring force and smaller local indentation.

During impact, the striker has greater displacement as smaller beam mass reduce the
overall mobility of weightless mass. Following the argument presented in pervious
comparison, for flexible beam case, maximum beam displacement in free vibration
stage is smaller due to small overall mobility of weightless mass.

In this parametric study, it is observed that the impact force is insensitive to flexural
stiffness of beam until it reaches an impractical circular frequency of 1000 Hz. It is
because inertia of beam contributes most of impact resistance. For short duration
impact pulse, a huge acceleration is required to change the velocity of striker in a very
short period of time. Inertial force, which is proportional to the acceleration of beam,

33
 Final Year Project: Structures Under Impact Loading

becomes very large also. On the other hand, the displacement of beam during impact
is very small. The spring force is very small also.

By increasing the stiffness of linear spring, the difference between beam

displacements at the end of impact is very small but the maximum displacement of
beam during free vibration decrease significantly. This observation indicates that the
flexural stiffness does not have significant contribution to beam impact behavior.

3.4 Comparison of impact models

There are three impact models present in this report: massive beam model, effective
mass model and beam vibration model. There are five parameters governing the
impact mechanics: mass of striker, ms, velocity of approach, vo , contact constant ,K,
mass of beam, mb and flexural stiffness k. Among them, three factors are related to
beam itself, they are local indentation, inertia and flexural stiffness.

Factors considered?
Models Local Flexural
Inertia
Indentation Stiffness
Massive beam model Yes No No
Effective mass model Yes Yes No
Beam vibration model Yes Yes Yes

The table above summarizes the factors considered in the model. Massive beam
model is clearly the upper bound solution of impact problems. Effective mass model,
which is shown by graphs below, often give a greater impact force compared with the
impact force obtained from beam vibration models. The duration of impact is also
shorter.

Mass Ratio α
Local Indentation Control Massive
Beam Model
Local Indentation Control
Inertia of Beam Control Effective
Mass Model

Beam
Vibration
Combined effect Model

Energy
Model
Flexural Stiffness
Control Pulse Ratio β

Energy model in this figure is a model which assumes all kinetic energy of striker is
transferred to elastic energy stored in beam flexure.

34
 Final Year Project: Structures Under Impact Loading

For massive and stiff beams, massive beam model can give a reasonable accurate
impact force history. It is because the beam is immovable if inertia of beam or flexural
stiffness is sufficiently large.

In another extreme, energy models yield an accurate result if the mass ratio is
sufficiently small. The kinetic energy stored in the beam can be negligible.

Different cases are plots below for comparison

Large
Reference Massive Flexible
Parameters Contact
case Beam Beam
Constant
1. Mass of striker , ms 1.5 kg 1.5 kg 1.5 kg 1.5 kg
2. Velocity of approach, vo 5 m/s 5 m/s 5 m/s 5 m/s
3. Contact Constant ,K 5 kN/mm3/2 5 kN/mm3/2 5 kN/mm3/2 20 kN/mm3/2
4. Mass of beam, me 1.5 kg 20 kg 1.5 kg 1.5 kg
5. First mode circular
300 Hz 300 Hz 15 Hz 300 Hz
frequency , ϖ1

Time History of Impact Force (Reference Case)

20000

Beam Vibration Model

Massive Beam Model
15000
Effective Mass Model
Impact Force(N)

10000

5000

0
0 0.2 0.4 0.6 0.8 1 1.2 1.4
Time (ms)

35
 Final Year Project: Structures Under Impact Loading

Time History of Impact Force (Massive Beam)

25000

Beam Vibration Model

20000
Massive Beam Model
Effective Mass Model
Impact Force(N)

15000

10000

5000

0
0 0.2 0.4 0.6 0.8 1 1.2 1.4
Time (ms)

Time History of Impact Force (Flexible Beam)

20000

18000
Beam Vibration Model
16000
Massive Beam Model
14000 Effective Mass Model
Impact Force(N)

12000

10000

8000

6000

4000

2000

0
0 0.2 0.4 0.6 0.8 1 1.2 1.4
Time (ms)

36
 Final Year Project: Structures Under Impact Loading

Time History of Impact Force (Large Contact Constant)

35000

30000 Beam Vibration Model

Massive Beam Model
25000 Effective Mass Model
Impact Force(N)

20000

15000

10000

5000

0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Time (ms)

37
 Final Year Project: Structures Under Impact Loading

CHAPTER FOUR

Experimental Investigation

Pin-ended simply supported perspex beams were tested by dropping a rigid striker
onto the beam at mid-span. Three sets of impact, named as flexible beam impact, stiff
beam impact and stiff and massive beam impact, were carried out to verify the beam
vibration model in impact with wide range of impact parameters. Acceleration and
maximum beam downward displacement at mid-span of beam during impact was
measured. The experiments were intended to verify the beam impact vibration when
the impact zone and the whole beam remain elastic throughout the whole process.

Some subsidiary experiments were also done to obtain essential parameters in beam
vibration model. 3 tensile tests were done to measure the Young modulus of perspex
beam. 3 static indentation tests were done to determine the relationship between
impact force and the local indentation.

The experimental result obtained is then compared with the theoretical prediction.

38
 Final Year Project: Structures Under Impact Loading

4.1 Experimental Setup

10 Acceleration History
11

Pull

9 Output
8
Data
Signal Logger
Amplifier
7
3 4 5
1 6

Instruments Description and Usage

1. Steel support Rigid steel with cylindrical contact surface with the beam
It is a device composed of a small perspex plate supported by 3
2. Displacement needles. These 3 needles are punched through a tube of
recorder plasticine. It recorded the maximum downward displacement
of beam vibration.
3. Perspex beam Beams with different dimensions for impact test
Accelerometers used are designed for impact experiments.
4. Accelerometer They are capable to record acceleration up to 10,000g.
( Kistler 8742A5 ) Accelerometers were attached on the beam surface near the
mid-span to record mid-span acceleration.
Signal amplifier was used to amplify the current from
5. Signal Amplifier
accelerometers to a voltage which is detectable by data logger.
Data logger having a sampling rate of 100kHz was used to
6. Data logger
capture the acceleration in a very duration of time
Ball joints were used to avoid uplift of beam at support during
7. Ball joint impact and free vibration stage. Ball joints can restrain the
upward translation without giving rotational restraints to beam.
A 3.6m long steel tube with 30mm internal diameter was used
8. Steel tube as the guide of striker. There was no welding on the steel tube
so a smooth passage of striker can be achieved.
Strikers were specially designed to reduce friction and ensure
9. Striker
accurate point of impact.
10. Pulley Cotton string was attached to the striker. Pulley and string
form a pulley system so that the striker can raise to desire
11. cotton string
height by pulling the string at ground level

39
 Final Year Project: Structures Under Impact Loading

4.2 Procedures of experiments

Acceleration history and maximum beam displacement are measured in separated

experiments. For experiment measuring the mid span beam acceleration, the
displacement recorder was removed to eliminate experimental error.
b
Measuring beam mid-span acceleration

1. Place the perspex beam on the support. Allow the striker resting on the beam
and then adjust the position of beam such that the point of contact with the
striker coincident with the centroid of beam.
2. Install the accelerometer onto the slot which is drilled before the experiment
3. Pull the string attached on the striker to the required height
4. Release the string to allow free falling on the beam
5. The data logger store the acceleration history at mid span of beam

Measuring maximum mid span beam downward displacement

1. Place the perspex beam on the support. Allow the striker resting on the beam
and then adjust the position of beam such that the point of contact with the
striker coincident with the centroid of beam.
2. Place the displacement recorder on a rigid platform at sufficient elevation such
that the beam can hit the displacement recorder during impact.
3. Pull the string attached on the striker to the required height
4. Release the string to allow free falling on the beam
5. The needles in the displacement recorder should be forced into the plasticine by
the beam downward displacement. Carefully pull the needles upwards for
several mm.
6. Place the displacement recorder to the original position.
7. Repeat step 3 and step 4
8. As shown below, the distance between the top of small perspex plate and the
bottom of perspex beam is measured. The measured distance is the maximum
beam displacement.

Maximum beam
displacement

40
 Final Year Project: Structures Under Impact Loading

4.3 Detailing of experiment apparatus

Two strikers, 1.5kg striker and 0.55kg striker, were specially designed for this
experiment. The figure below shows the configuration of the striker.

Metal hoop

Small screw
20mm
CL
Small spring

Steel rod

CL
Small hole was 45mm
drilled through
the rod
φ 16 steel
sphere

The above striker was specially design to achieve accurate point of impact and
reducing the friction loss during impact test. The φ16mm steel sphere is welded on the
steel rod to provide spherical contact surface which matches the assumption in the
Hertz Contact Law. As shown above, a small spring and screw was placed into a small
hole drilled in the metal rod. Small compression force has to be applied on spring to
put the striker inside the steel tube. This arrangement not only reduces the contact
between the striker and the steel tube, but also makes sure the centroid of striker
follow the centerline of tube throughout the impact.

41
 Final Year Project: Structures Under Impact Loading

4.4 Dimensions of experiments setup

CL

L’ L’
h’

Flexible beam impact Stiff beam impact Massive beam impact

L’ 50 mm 25 mm 25 mm
h’ 25 mm 25 mm 25 mm
s 50 mm 25 mm 25 mm

A small slot has to be drilled for fixing the accelerometers. 25mm distance between
the slot and the mid span of beam is specified to make sure the slot has negligible
influence to the stress distribution at impact zone. In flexible beam, the beam
displacement is much greater. Horizontal displacement was observed after the impact.
Large separation between the point of impact and accelerometer s is specified to avoid
the striker hit the accelerometer.

4.5 Configuration of beam impact test

Test variables required in beam vibration model are obtained as follow:

Parameters Description
Striker mass , ms Determined by direct measurement
By using formula: v0 = 0.9 2 gH , where H is the drop
Approach velocity vo distance. Correction factor 0.9, which is suggested by
some literature, accounts for the friction loss of strikers.
No calibration was done to determine the correction factor.
Contact Constant K From curve fitting of indentation test result
In the beam vibration model, the contribution due to
overhang is assumed to be negligible. Therefore, the
Beam mass mb weight of beams measured is reduced proportionally for
beam mass in the model. (i.e., mass of beam in flexible
beam impact = 15/16 of the mass of beam measured)
π2 EI
By analytical formula, ω =
2
, where the young
Beam frequency, ϖ1 L mb L2

modulus E was determined by tensile tests.

42
 Final Year Project: Structures Under Impact Loading

The table below summarizes all the measured values:

Flexible Beam Massive Beam

Stiff Beam Impact
Impact Impact
Dimensions
1600 x 33 x 20 550 x 100 x 20 550 x 100 x 50
(B x D x L)
Span length 1.5 m 0.5 m 0.5 m
Mass of striker ms 1.5 kg 0.55 kg 0.55 kg
Mass of beam, mb 1.17 kg 1.17 kg 2.71 kg
Natural period t1 154.4 ms 17.1 ms 6.665 ms
Height of impact H 1.5 m 0.5m 3m

The non-dimension parameters governing the impact behavior are listed below:

Flexible Beam Massive Beam

Stiff Beam Impact
Impact Impact
Mass ratio α 0.78 2.13 4.93
Pulse ratio β 0.00848 0.0430 0.0923

Therefore, the experiment attempts to verify the beam vibration model for a wide
range of parameters.

4.6 Experimental results

4.6.1 Tensile test result

Young Modulus measured

Sample 1 3041 MPa
Sample 2 2906 MPa
Sample 3 2952 MPa

The Young Modulus of the perspex is taken as 2968 MPa which is the average of 3
tensile test results.

4.6.2 Indentation Test result

Experiments have carried out to measure the force indentation relationship. From
curve fitting of experimental result in the form of F = Ka n , the following are the
result used for beam vibration analysis:

Force Indentation relationship

Sample from flexible beam F = 6.61a 1.49
Sample from rigid beam F = 5.67 a1.62
Sample from massive beam F = 3.9a1.78
where F is in kN and a is in mm

43
 Final Year Project: Structures Under Impact Loading

4.6.3 Maximum Beam Displacement

Beam Vibration Model Experimental Value

Flexible Beam Impact 68 mm* 102 mm
Stiff Beam Impact 4.22 mm 3 mm
Massive Beam Impact 3.89 mm 2 mm
* - Only two subsequent impacts has considered

4.6.4 Mid-span acceleration

All the beam impact experiments presented in this report have been repeated for
several times to check against the consistency of the experimental results. 3 identical
“flexible beams”, 3 identical “stiff beams” and 2 identical “massive beams” are tested.
Results obtained were compared also. It is found that the results obtained are very
consistent. They differ about only 5-10%.

5% damping ratio was incorporated in the beam vibration model. As the beam
undergoes free vibration, the effect of damping becomes more significant, especially
for higher vibration modes. Damping can be clearly identified in the experimental
data collected also. The 5% damping ratio is commonly used to account for damping
of structure in elastic stage. In this experimental study, this damping ratio was adopted
without any experimental verification.

The following graphs show the mid span acceleration of “Flexible Beam Impact”,
“Stiff Beam Impact” and “Massive beam impact”. More experimental result can refer
to Appendix D.

Mid-span Acceleration for Flexible Beam Impact

40000

Experiment Result
30000
Beam Vibratoin Model
Beam Vibratoin Model (E+50%)
20000
Acceleration (m/s 2)

10000

0
0 0.002 0.004 0.006 0.008 0.01 0.012

-10000

-20000
Time (s)

44
 Final Year Project: Structures Under Impact Loading

Mid-span Acceleration for Stiff Beam Impact

15000 Channel 1
Beam Vibraion Model
Channel 2
10000 Beam Vibraion Mode(E+50%)l
Acceleration (m/s2)

5000

0
0 0.0002 0.0004 0.0006 0.0008 0.001 0.0012 0.0014 0.0016 0.0018 0.002

-5000

-10000
Time (s)

Mid-span Acceleration for Massive Beam Impact

30000 Channel 1
Beam Vibration Model
25000
Channel 2
20000 Beam Vibration Model (E+50%)

15000
Acceleration (m/s2)

10000

5000

0
0 0.0005 0.001 0.0015 0.002 0.0025 0.003 0.0035 0.004
-5000

-10000

-15000

-20000
Time (s)

4.7 Discussion of experimental results

Generally, the beam vibration model can capture the trend of acceleration quite well
in all three impact scenarios.

From the above graphs showing mid-span acceleration against time, it is observed that
the beam vibration model gives a good approximation to “Stiff Beam Impact” and
“Massive Beam Impact”. However, the peak cannot be modeled accurately in
“Flexible Beam Impact”. Also, the beam in flexible beam impact seem to softer in
45
 Final Year Project: Structures Under Impact Loading

actual case because the acceleration measured is smaller. For stiff beam impact and
massive beam impact, the acceleration measured is greater than the predication by
beam vibration model. Hence, the beam in stiff beam impact and massive beam
impact behave more rigid in the actual case.

The reason for inaccurate modeling of flexible impact because the pulse ratio β is
very small ( β =0.00848). The transverse shear wave speed:

G E 2968 × 10 6
vs = = = = 979m / s ………………….(?)
ρ 2(1 + υ )ρ 2(1 + 0.3) × 1192

The shear wave requires 0.766ms to travel from the mid-span to the support if span
length is 1.5m. The time required is in the same order of magnitude of the impact
duration. It must be recognized that beam vibration model describe the beam behavior
by assuming the impulse waves travel with an infinite velocity. This assumption is no
longer valid in flexible beam case.

In all the experiments, it is observed that the periods of vibration measured in all three
impact experiments are shorter than prediction done by beam vibration model. By
increasing the Young Modulus by 50% in all three impact cases, the beam vibration
model gives a much better match with the experimental results. One of the possible
reasons for this observation is the enhancement in Young Modulus due to strain rate
sensitivity which is not included in the beam vibration model. My literatures have
recognized that Young Modulus is a function of strain rate rather than a constant for
high strain rate.

For stiff beam impact, a very interested phenomenon has observed form the graph
below:
Mid-span Acceleration for Stiff Beam Impact
60000 Channel 1
Beam Vibraion Model
Channel 2
40000
Beam Vibraion Model (E+50%)

20000
Acceleration (m/s2)

0
0 0.001 0.002 0.003 0.004 0.005 0.006

-20000

-40000

-60000
Time (s)

46
 Final Year Project: Structures Under Impact Loading

Mid-span Velocity for Stiff Beam Impact

4
CH1
3 CH2
Beam Vibration Model
2 Beam Vibration Model (E+50%)
Velocity (m/s)

0
0 0.001 0.002 0.003 0.004 0.005 0.006

-1

-2

-3
Time (s)

The mid-span acceleration graph shown below was obtained from extending the time
axis from 0.002s to 0.006s from acceleration graph shown in previously graph. At
time 0.002s, high frequency waves with very high acceleration superimpose onto the
original graph. By inspection, the frequencies of incident waves are between 14-16
kHz. By plotting mid-span velocity graphs, we can see that the incident waves have
little effect on the mid-span velocity. The original trend can be maintained in the
velocity-time graph. As high frequency waves have high acceleration but low velocity,
we can conclude that the incident waves are mainly composed of high frequency
waves.

1. When the striker hit the 2. At the impact zone the 3. The impact zone rebound
beam , instead of shear striker and the beam upwards as a result of
transverse waves, temporary lose contact relaxation of longitudinal
compressive longitudinal and the longitudinal waves. The striker hits the
waves also generated waves travel downward beam again and
longitudinal waves
generated again

Resonance may occur if the frequency of wave generated is closed to the resonant
frequency of beam due to its thickness.

47
 Final Year Project: Structures Under Impact Loading

From the calculation below,

E 2968 × 10 6
vp = = = 1578m / s ………………….(?)
ρ 1192

vp 1578
fp = = = 19.7 kH z ………………….(?)
4H 4 × 0.02

where H is the thickness of the beam

The resonant frequency is close to the frequency of incident waves. Therefore, the
resonance frequency of longitudinal waves is a possible reason for this observation.

The massive beam impact has some incident high frequency waves of 6 – 8 kHz at
above 1.3ms. The resonance frequency of longitudinal waves is 7.9 kHz which also
support the above argument.

48
 Final Year Project: Structures Under Impact Loading

CHAPTER FIVE

Application of Beam Vibration Model

The philosophy of beam vibration model is to employ separated models for the local
behavior of impact zone and global behavior of structure and then employ some
mechanism to couple these two models. In real-life application, the local behavior of
impact zone is often refers to localized damage or inelastic deformation at point of
impact. Global behaviors refer to overall structural vibration induced by dynamics
impact loading. In this project, Hertz law is used to model non-linear behavior at
impact zone. Undamped MDOF dynamics system is employed to model the structural
vibration of simply supported beam.

Very often, the local behaviors at impact zone are extremely complicated. Non-linear
deformation of both striker and impact zone, stress distribution within impact zone
and mechanism of fracture make the accurate modeling of local behavior extremely
difficult. On the other hand, the global behaviors of structure are more predictable.
Very often, the structures as a whole can remain elastic but it suffers heavy damage at
impact zone. Therefore, the local behavior and global behavior should be treated
separated.

49
 Final Year Project: Structures Under Impact Loading

Detail assessment of Coupling 2 Global and

local behavior (e.g. models using local
non-linear Finite beam vibration response of
Element Analysis) Force and local model structure
deformation design
curves

Using well-developed
structural dynamics analysis

Global dynamics
parameters (e.g.
effective masses and
periods)

In the modeling of the local behavior of impact zone, the impact zone should be
defined such that damages and complex stress distribution of concentrated impact
force can be confined in the impact zone. In modeling of the impact zone, the
boundary conditions applied on the impact zone should be fully restraint in both
translation and rotation. Detail assessments such as shock resistance of construction
material are considered to derive the relationship between force and local indentation.
Empirical design curves can also be produced for different impact parameters.

For the global behavior of structure, only the global dynamics parameters such of
modal mass, periods for different vibration modes are required for the modeling. The
beam vibration models present in this report can couple these two models for a more
realistic global and local behavior of structures under impact.

50
 Final Year Project: Structures Under Impact Loading

Failure by local damage to

structural core
(Local Response)

Failure by excessive
structural vibration
(Global Response)

The global behavior is concerned if the structure is flexible, light and its impact zone
is rigid. The structural vibration induced by the impact may be so great that the
structure cannot be sustained. On the other hand, if the structure is very massive and
its impact zone is ductile. The structure may suffer from heavy localized damage with
small overall structural vibration. The major structural members at the impact zone
may be heavily damaged. If the gravity load redistribution to other members is
insufficient, the structure may collapse due to failure of vertical loading carrying
system.

51
 Final Year Project: Structures Under Impact Loading

CHAPTER SIX

Conclusions & Suggestions

6.1 Theory Studies

1. The beam impact problem has been addressed by the use of three models to
describe the beam behavior. The first model considered the beam as massive
compared with the striker and thus exhibiting no displacement. The second model
considered the beam as the effective concentrated mass, and made some
allowance for beam displacement. The third model described the beam
displacement more accurately by consideration of high modes beam vibration.

2. In beam vibration model, Hertz contact law models the local indentation and
MDOF undamped dynamics system models the global beam vibration. These two
models are coupled by compatibility equation. Subsequence impact can also be
modeled by introducing impact criteria in beam vibration model.

3. Three non-dimensional parameters, mass ratio α, pulse ratio β and

non-dimensional impact force F’ are identified. Parametric analysis has carried
out to investigate how the above parameters govern impact force history, local
indentation and beam vibration. An intuitive analogy - two spring model is
suggested and used to explain the findings in parametric analysis.

4. Generally, high mass ratio results in greater impact force and longer impact
duration. High pulse ratio means higher vibration modes are likely to be excited
during impact. Larger contact constant result in increase in impact force but
reduction in impact duration.

5. Different simpler models (massive beam model, effective mass mode and energy
models) can gives good approximation in certain range of mass ratio α and pulse
ratio β . The boundaries of these parameters determining whether simpler models

52
 Final Year Project: Structures Under Impact Loading

are applicable have not been identified yet.

6. For very stiff impacts, the beam vibration model shown to be inadequate, and a
correction solution would required a wave-travel solution.

6.2 Experimental Studies

1. Generally speaking, the experimental result shows good agreement with the beam
vibration model for a wide range impact parameters α and β .

2. The beam vibration model gives more accurate solution for stiff beam impact and
massive beam impact. Larger errors were found in flexible beam impact (small β
value) because the assumption of infinite wave speed is not valid.

3. Strain rate sensitivity has significant influence in beam impact test. Beams seem
to behave stiffer under impact. Simply increasing the young modulus by 50%
yields a much better result in beam vibration analysis. However, free vibration
tests should be performed to further verify this observation.

4. Impact force was not measured in experiments. Load cells should be attached to
the striker to verify the impact force history obtained from beam vibration analysis

5. Some high frequency waves were measured in some impact tests. These high
frequency waves cannot be predicated by beam vibration model. Preliminary
investigation indicates that the source comes from resonance of longitudinal
waves. Further studies are required to identity the source of such high frequency
wave.

6. The loading and unloading curves in indentation tests are not identical. Allowance
should be made to cater for energy lost in loading and unloading process.

6.2 Application

1. Beam vibration model can apply for a wide range of impact problems. For
complicated structure, method of modeling for impact loading was given. It is
done by treating the impact zone and global structure behavior separately and
coupling them again by beam vibration model.

53
 Final Year Project: Structures Under Impact Loading

Reference

1. G. Huges & D.M. Speirs , An Investigation of the beam impact problem, Cement
and concrete Association , 1983

2. W.J. Stronge , Impact Mechanics , Cambridge University Press , 2000

3. Ray W. Clough, Joseph Penzien , Dynamics of Structures second edition,

McGraw Hill , 1975

4. G. S. T. Armer & F.K. Garas , Design for Dynamic Loading, the Use of Modal
Analysis , Longman Inc., New York , 1982

5. T. A. Laursen , Computational Contact and Impact Mechanics , Fundamentals of

Modeling Interfacial Phenomena in Non-linear Finite Element Analysis ,
Springer , 2002

6. N. Joes, D.G. Talaslidis, C.A. Brebbia, G..D. Manolis , Structure under Shocks
and Impact V , Computational Mechanics Publication , 1998

7. C. Gibbons , Safety in Tall Buildings and other Building of Large Occupancy ,

Ove Arup & Partners

54
Appendix A

The Numerical Solution of the impact

equation
 2/ 3 ∞
 F (t )  1 t τ 2 1
∑ ω ∫ F (τ ) sin ω (t − τ )dτ
t
  = vot − ∫∫ F (t )d tdτ − i
 K  ms 0 0 mb i =1,3 i
0

………………….(2-42)

Equation (2-42) in main report is the governing equation of beam vibration model.
This equation cannot be solved analytically. Numerical technique has to be employed
to obtain the force history of impact force. The numerical solution used in this project
is modified from numerical model suggested by Dr R Sym, from the Operation
Research Development of the C&CA, is used.

The following notation is used in this appendix.

∆t length of time interval

ti = i∆t, time at the end of i th time interval
ci = cos (ϖi∆t)
si = sin (ϖi∆t)
Fi impact force at the end of i th time interval
K
Gk = ∑ Fi , sum of forces up to time tk
i =1

The following are the initial condition in beam impact problem

t0 = F0 = G0 = 0 , s0 = 0 and c0 = 1

The numerical solution is obtained by the following assumption

1. The infinite summation over vibration modes is replaced by a finite summation to

the first N vibration modes.
2. The integration with respect to time are replaced by summations over a finite
number of time intervals

The first approximation is straightforward. Number of vibration modes included in the

solution should be determined according to the physical solution being examined. In
this report, 29 vibration modes are included in all the beam vibration analysis.

To achieve second approximation, the impact duration is divided into a number of

equal time interval of length ∆t. and the impact force, F(t), between the striker and the
beam is assumed to vary linearly within each time interval. This allows integrations to
be carried out analytically, so that the integrals in equation (2-42) can be replaced by
summations. The approximations can be improved by reducing the size of the time
intervals used. Choosing a ∆t such that about 10 intervals occurred up to the
maximum impact force was found to give reasonable accurate result.

Replacing a function, such as F(t), by a piecewise linear approximation is a traditional

method of numerical integration. However, the impact force in the equation (2-42) is
the unknown. At the start of impact (i.e. t =0), the force is F(0) = F0 = 0. At t=∆t, the
approximation for the equation involves the force F(∆t) = F1 at the end of time
interval as the only unknown. Hence, at time t = ∆t, the equation can be solved for F1.
Similarly, if later time interval , say m∆t to (m+1)∆t , is considered, if the impact
force history F(t) is known up to and including time t=m∆t, the approximation for
equation (?) then involves the force Fm+1 as the only unknown. Hence the force
history ( F1, F2, F3, …) can be calculated by solving the approximation for equation
(2-42) at the succession of time t=∆t , 2∆t , 3∆t , …

The following sections of this Appendix deal with the linear approximation formula
for the force history, the derivation of recurrence relations for the integrals in equation
(2-42), and the solution of the approximation for this equation for the impact force at
a particular time if the previous force history is given.

For the ith time interval, i.e. for

t i −1 ≤ t ≤ t i

the linear approximation is given by

t − t i −1
F (t ) = Fi −1 + (Fi − Fi −1 ) ………………….(A-1)
t i − t i −1

Recurrence relation for the displacement of the striker

In equation (2-42), the terms

1 t τ
vo t −
ms ∫ ∫ F (t)d tdτ
0 0
………………….(A-2)

Represent the displacement of the striker. At a time tm , we define

t τ
Dm = ∫ ∫ F (t)d tdτ
0 0
………………….(A-3)

It is shown below that the following recurrence relation holds for Dm for m = 0,1,2,…

 F − Fm 
Dm+1 = Dm + (∆t ) 2 Gm + m +1  ………………….(A-4)
 6 

m
where Gm = ∑ Fi by definition
i =1
Noted that D0 = 0 ,and that Gm can also be calculated using a recurrence relation, for
m = 0,1,2,…

Gm +1 = Gm + Fm +1 ………………….(A-5)
Hence the terms in equation (A-2) which represent the displacement of the striker can
be written as

v0 (m + 1)∆t −
1
Dm +1 ………………….(A-6)
ms

To derive the recurrence relation, the first stage is to substitute equation (A-1) in
equation (A-3), and carry out integration. This gives

 m  k −1  F 
Dm = (∆t ) ∑  ∑ Fj  + m 
2
………………….(A-7)
 k =1  j =1  6 

Using equation (A-7), and the definition of Gm, one can then calculate that

2 F − Fm 
Dm+1 − Dm = (∆t ) Gm + m+1 
 6 

which is the same as equation (A-4) , the recurrence relation for Dm.

Pin-ended beam impact – recurrence relations for beam displacement

In equation (2-42), the term

∞
1 1
∑ ω ∫ F (τ ) sinω (t − τ )dτ
t
i ………………….(A-8)
mb i =1,3 i
0

represents the beam displacement. At a time tm during the course of impact, define

tm
S i,m = ∫ F (τ ) sin ωi (t − τ )dτ ………………….(A-9)
0

and

tm
Ci ,m = ∫ F (τ ) cosωi (t − τ )dτ ………………….(A-10)
0

Noted that the subscript i, which is required here to distinguish between vibration
modes and which is used only for this purpose, does not enter into the derivation of
the recurrence relations. It is shown below that the following recurrence relations hold
for Si,m and Ci,m , for m = 0,1,2,…

 s  F − Fm (1 − ci )Fm
S i,m = ci Si ,m + si Ci ,m + 1 − i  m+1 + ………………….(A-11)
 ωi ∆t  ωi ωi
 1 − ci  Fm+1 − Fm si Fm
Ci ,m = ci Ci,m + si S i,m +   + ………………….(A-12)
 ωi ∆t  ωi ωi

Note that Si,0 = Ci,m = 0 as there is no impact force at t = 0 . The term in equation
(2-42) which represents the displacement of the beam can be written, using equation
(A-11).

∞
1 2
mb
∑ω
i =1,3
S i,m+1 ………………….(A-13)
i

At time tm+1, the force history enters only through terms which can be calculated using
recurrence relations. Equation (A-9) and (A-10) are both required because the
calculation of a sine term Si,m+1 requires both of the previous sine and cosine terms
Si,m and Ci,m .

In equation (A-11) and (A-12) the terms in square brackets involved division by ∆t.
This need not cause numerical problem if ∆t is made very small because, in either
case, the term as a whole approaches zero as ∆t is small.

The recurrence relation for Si,m can be obtained by taking the definition of Si,m, i.e

t m +1
S i,m+1 = ∫ F (τ ) sin ωi (t − τ )dτ ………………….(A-14)
0

and breaking the integration into 2 ranges, 0 to tm and tm to tm+1. Substituting the
trigonometric relationship

sin{ωi (t m+1 − τ )} = s1 cos{ωi (t m − τ )} + c1 sin{ωi (t m − τ )} ………………….(A-15)

into the first range gives

tm
∫ F (τ ) sin ωi (t − τ )dτ = si Ci ,m + ci S i,m ………………….(A-16)
0

Substituting equation (A-15) into the second range, and integrating by parts, gives

t m +1  s  F − Fm (1 − ci )Fm
∫ F (τ ) sin ωi (t − τ )dτ = 1 − i  m+1 + ……………….(A-17)
tm
 ωi ∆t  ωi ωi

Equation (A-16) and (A-17) together gives equation (A-11). Equation (A-12) can be
derived similarly. These recurrence relations provide a convenient way of computing
solutions of equation (2-42).
Pin-ended beam impact – calculation of the force history

At a time tm+1 the approximation of equation (2-42) contains the unknown Fm+1 in
both sides of the equation, and the known force history ( F1, F2, … , Fm) in terms
which can be calculated using recurrence relations. Because the left hand side
contains the fractional power 2/3 , the equation has to be solved by an iterative
method. The approximation to equation (2-42) can be written, at time tm+1,

2/3
 Fm+1 
  = Am + BFm+1 ………………….(A-18)
 K 
where, in pinned case,

2 F    si 
Dm + (∆t )  Gm − m   F  − c  
 6  1 N
2 
m
 ω i ∆t i

Am = vo (m + 1)∆t −
ms
−
mb
∑
i =1,3 ωi

ci S i ,m + s i Ci ,m +
ωi 
 
 
…..….(A-19)
and
(∆t )
2
1 N
2  s i ,1 
B=−
6m s
−
mb
∑ω
i =1,3
2 
i 
1−
ω i ∆t 
 ………………….(A-20)

Note that B does not vary with time. Also, it can be seen that B is negative, because all
the physical constants in it are positive and the term in curly brackets can be shown to
be positive: hence equation (A-18) has a positive solution, which is unique, if Am is
positive, and it has no solution if Am is negative. During the calculations, it is found
that Am remains positive until the force become very small near the end of the impact.
Appendix B

The Numerical Solution of the impact

equation
A. Varying contact constant K

Parameters Values
1. Mass of striker , ms 1.5 kg
2. Velocity of approach, vo 5 m/s
3. Contact Constant ,K 2 – 20 kN/mm3/2
4. Mass of beam, mb 1.5 kg
5. First mode circular frequency , ϖ1 300 Hz

The following graphs are plotted below.

Comparsion of Impact Force History when beam mass and stiffness are fixed

11
10 K=2kN/mm^3/2
9 K=5kN/mm^3/2
8 K=10kN/mm^3/2
Impact Force (kN)

7 K=20kN/mm^3/2
6
5
4
3
2
1
0
0 0.5 1 1.5 2 2.5
Time (ms)

Comparsion of beam vibration when beam mass and stiffness are fixed

0.025
K=2kN/mm^3/2
0.02
K=5kN/mm^3/2
0.015
K=10kN/mm^3/2
0.01 K=20kN/mm^3/2
Displacement (m)

0.005
0
0 5 10 15 20 25
-0.005
-0.01

-0.015
-0.02
-0.025
Time (ms)
 Comparsion Local Indentation and Max beam displacement

25.00
Max Local indentation

20.00 20.02 Max Beam displacement

Beam Vibration at the end of impact
Displacement (mm)

15.98
15.00 14.68 15.08

10.00
7.54
5.00 4.43 3.85
1.76 2.05
1.15 0.87
0.00 0.65
0 2 4 6 8 10 12 14 16 18 20
Contact Constant (kN/mm3/2)

B. Varying beam mass mb

Parameters Values
1. Mass of striker , ms 1.5 kg
2. Velocity of approach, vo 5 m/s
3. Contact Constant ,K 5 kN/mm3/2
4. Mass of beam, mb 0.5 - 10 kg
5. First mode circular frequency , ϖ1 300 Hz

The following graphs are plotted below.

Comparsion of Impact Force History when contact constant and beam stiffness are fixed
15
Me=0.5kg
12.5 Me=1.5kg
Me=5kg
Impact Force (kN)

10 Me=10kg

7.5

2.5

0
0 0.2 0.4 0.6 0.8 1 1.2 1.4
Time (ms)
 Comparsion of Beam Displacement when contact constant and beam stiffness are fixed
0.02 Me=0.5kg
Me=1.5kg
0.015
Me=5kg
0.01 Me=10kg
Displacement (m)

0.005

0
0 5 10 15 20 25
-0.005

-0.01

-0.015

-0.02
Time (ms)

Comparsion Local Indentation and Max Beam Displacement

20.00
18.10 Max Local indentation

16.00 Max Beam displacement

14.68 Beam Vibration at the end of impact
Displacement (mm)

12.00
10.95

8.00
7.34

4.71 4.43
4.00
2.95 2.03
1.72
0.76 1.15 1.99
0.00
0 1 2 3 4 5 6 7 8 9 10
Effective Mass of Beam (kg)
 C. Varying flexural stiffness of beam k

As the mass of beam is fixed, increasing the flexural stiffness is equivalent to

increasing the first mode circular frequency ϖ1 according to this fundamental
formula:

k1 = meω12 ………………….(?)

The flexural stiffness k is approximated by first vibration mode stiffness k1

Parameters Values
1. Mass of striker , ms 1.5 kg
2. Velocity of approach, vo 5 m/s
3. Contact Constant ,K 5 kN/mm3/2
4. Mass of beam, me 1.5 kg
5. First mode circular frequency , ϖ1 15 - 1000 Hz

Comparsion of Impact Force History when Contact Constant and beam mass are fixed
10
Freq=15Hz
Freq=50Hz
7.5
Freq=150Hz
Freq=300Hz
Impact Force (kN)

5 Freq=1000Hz

2.5

0
0 0.2 0.4 0.6 0.8 1 1.2 1.4

-2.5
Time (ms)
 Comparsion of Beam Displacement when Contact Constant and beam mass are fixed
0.25

0.2

0.15 Freq=15Hz
Displacement (m)

Freq=50Hz
0.1 Freq=100Hz
Freq=300Hz
0.05 Freq=1000Hz

0
0 20 40 60 80 100 120
-0.05

-0.1
Time (ms)

Comparsion Local Indentation and Max beam displacement

219 for ϖ1 = 15Hz
70.00 Max Local indentation
66.31
60.00 Max Beam displacement
Beam Vibration at the end of impact
Displacement (mm)

50.00

40.00

30.00
23.70
20.00
15.98
10.00 3.15 6.74
3.11 3.40 4.43
3.40
0.00 1.41
1.12 1.12 1.13 1.15
0 100 200 300 400 500 600 700 800 900 1000
Circular Frequency of 1st Vibration Mode (Hz)
Appendix C

Photos of Experimental Setup

General View of experimental Setup

Beam used for impact Test

 Impact Testing for Massive Beam

Accelerometers installed onto stiff beam

0.55 kg Striker

1.5 kg Striker
Accelerometers KISTLER 8742A5and corresponding Signals amplifiers

Accelerometers B&K 4370 and corresponding Signals amplifiers

 Ball Joint

Displacement recorders
Sampler for indentation test

Indentation test
Appendix E

Graphs for Beam Impact Test

 Flexible Beam Impact Test – Charts

Mid-span Acceleration for Flexible Beam Impact

7
Experimental Result
6
Beam Vibratoin Model

5 Beam Vibratoin Model (E+50%)

4
Velocity (m/s)

0
0 0.002 0.004 0.006 0.008 0.01 0.012 0.014 0.016 0.018 0.02
-1

-2
Time (s)

Mid-span Velocity for Flexible Beam Impact

7
Experimental Result
6
Beam Vibratoin Model

5 Beam Vibratoin Model (E+50%)

4
Velocity (m/s)

0
0 0.002 0.004 0.006 0.008 0.01 0.012 0.014 0.016 0.018 0.02
-1

-2
Time (s)
 Mid-span Displacement for Flexible Beam Impact
0.08

0.07
Displacement(m)
0.06
Beam Vibratoin Model
0.05 Beam Vibratoin Mode (E+50%)l
Velocity (m/s)

0.04

0.03

0.02

0.01

0
0 0.005 0.01 0.015 0.02 0.025 0.03
-0.01
Time (s)

Stiff Beam Impact Test - Charts

Mid-span Acceleration for Stiff Beam Impact

60000 Channel 1
Beam Vibraion Model
Channel 2
40000
Beam Vibraion Model (E+50%)

20000
Acceleration (m/s2)

0
0 0.001 0.002 0.003 0.004 0.005 0.006

-20000

-40000

-60000
Time (s)
 Mid-span Velocity for Stiff Beam Impact
4
CH1
3 CH2
Beam Vibration Model
2 Beam Vibration Model (E+50%)
Velocity (m/s )

0
0 0.001 0.002 0.003 0.004 0.005 0.006
-1

-2

-3
Time (s)

Mid-span Displacement for Stiff Beam Impact

CH1
0.005 CH2
Beam Vibration Model
Beam Vibration Model (E+50%)
0.003
D is plac em ent (m )

0.001

0 0.002 0.004 0.006 0.008 0.01 0.012

-0.001

-0.003

-0.005
Time (s)
 Massive Beam Impact Test - Charts

Mid-span Acceleration for Massive Beam Impact

30000 Channel 1
Beam Vibration Model
25000
Channel 2
20000 Beam Vibration Model (E+50%)

15000
Acceleration (m/s2)

10000

5000

0
0 0.0005 0.001 0.0015 0.002 0.0025 0.003 0.0035 0.004
-5000

-10000

-15000

-20000
Time (s)

Mid-span Velocity for massive Beam Impact

4 CH1

3 CH2
Beam Vibration Model
2
Beam Vibration Model (E+50%)

1
Velocity (m/s)

0
0 0.0005 0.001 0.0015 0.002 0.0025 0.003 0.0035 0.004 0.0045 0.005
-1

-2

-3

-4

-5
Time (s)
 Mid-span Displacement for Massive Beam Impact
0.005
CH1
0.004 CH2
Beam Vibration Model
0.003 Beam Vibration Model (E+50%)
Displacement (m)

0.002

0.001

0
0 0.0005 0.001 0.0015 0.002 0.0025 0.003 0.0035 0.004 0.0045 0.005
-0.001

-0.002

-0.003
Time (s)
Appendix F

Graphs for Tensile Test

 Sample One Tensile Test - Stress Strength Curve
30

25
y = 3041.7x
20
Stress (MPa)

15

10

0
0 0.001 0.002 0.003 0.004 0.005 0.006 0.007 0.008 0.009
Strain

Sample 2 Tensile Test - Stress Strain Curve

30

25
y = 2906.1x - 0.0376
20
Stress (MPa)

15

10

0
0 0.001 0.002 0.003 0.004 0.005 0.006 0.007 0.008 0.009
Strain
 Sample 3 Tensile Test - Stress Strain Curve
50

45

40

35

30
Stress (MPa)

25
y = 2925x + 0.3392
20

15

10

0
0 0.002 0.004 0.006 0.008 0.01 0.012 0.014 0.016 0.018 0.02
Strain
Appendix F

Graphs for Indentation Test

 Loading & Unloading Curve for indentation test of flexible beam
16

14

12
Indentation Force (kN)

10

6
Loading
4
Unloading
2

0
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
Indentation (mm)

Loading & Unloading Curve for indentation Test of Stiff Beam

35

30

25
Loading
Impact Force (kN)

Unloading
20

15

10

0
0 0.5 1 1.5 2 2.5 3 3.5
Indentation (mm)
 Loading & Unloading Curve for indentation Test of Massive Beam

35

30

25 Loading
Impact Force (kN)

Unloading
20

15

10

0
0 0.5 1 1.5 2 2.5 3 3.5
Indentation (mm)

Curve Fitting Result of Indentation Test

35

30

25
F = 3.9a1.78
Indentation Force (kN)

20
F = 5.67 a1.62
15 F = 6.61a1.49
Flexible beam
10
Stiff Beam
Massive Beam
5

0
0 0.5 1 1.5 2 2.5 3 3.5
Indentation (mm)

View publication stats