Available

Available online
online at
at www.sciencedirect.com
www.sciencedirect.com

ScienceDirect
Available online at www.sciencedirect.com
Procedia
Procedia Engineering
Engineering 00
00 (2017)
(2017) 000–000
000–000
www.elsevier.com/locate/procedia
ScienceDirect www.elsevier.com/locate/procedia

Procedia Engineering 199 (2017) 3422–3427

X International Conference on Structural Dynamics, EURODYN 2017

On adjusting the rotary inertia of a cantilever-type energy harvester

for wideband operation
Peter Ibrahimaa, Omar Nassarbb, Mustafa Arafac,c,*, Yasser Anisaa
aa
Department
Department ofof Mechanical
Mechanical Design
Design and
and Production,
Production, Faculty
Faculty of
of Engineering,
Engineering, Cairo
Cairo University,
University, Egypt
Egypt
bMechatronics Engineering Department, Ain Shams University, Cairo, Egypt
b
Mechatronics Engineering Department, Ain Shams University, Cairo, Egypt
cc
Mechanical
Mechanical Engineering
Engineering Department,
Department, American
American University
University in
in Cairo,
Cairo, Egypt
Egypt

Abstract
Abstract

Inertial
Inertial vibration
vibration energy
energy harvesters
harvesters are are effective
effective only
only when
when they
they operate
operate at at their
their natural
natural frequencies.
frequencies. A A slight
slight deviation
deviation from
from
operation
operation at resonance conditions leads to a substantial reduction in the generated power, which places hurdles for the development
at resonance conditions leads to a substantial reduction in the generated power, which places hurdles for the development
of
of devices
devices that
that operate
operate across
across aa wide
wide frequency
frequency range.
range. ToTo address
address this
this issue,
issue, wewe present
present aa vibration
vibration energy
energy harvester
harvester with
with aa tunable
tunable
natural
natural frequency.
frequency. Frequency
Frequency adjustment
adjustment in in aa cantilever-type
cantilever-type energy
energy harvester
harvester is is achieved
achieved byby varying
varying the
the rotary
rotary inertia
inertia of
of the
the beam
beam
through
through changing
changing the the angular
angular position
position of of aa rotatable
rotatable mass
mass mounted
mounted at at its
its tip.
tip. The
The mass
mass isis driven
driven by
by aa beam-mounted
beam-mounted servomotor
servomotor
which
which controls
controls itsits angular
angular position
position about
about an an axis
axis normal
normal toto the
the beam’s
beam’s mid-plane,
mid-plane, thereby
thereby varying
varying the
the effective
effective mass
mass moment
moment ofof
inertia.
inertia. In
In this
this arrangement,
arrangement, the the power
power required
required for for frequency
frequency adjustment
adjustment is is kept
kept to
to aa minimum,
minimum, since
since nono mechanical
mechanical straining
straining
forces
forces are
are applied
applied on on the
the beam,
beam, andand power
power is is only
only needed
needed when
when thethe motor
motor turns
turns to
to adjust
adjust the
the orientation
orientation ofof the
the tuning
tuning mass.
mass. AA
frequency
frequency tuning
tuning algorithm
algorithm isis developed
developed to to keep
keep the
the beam
beam operating
operating atat resonance.
resonance. The The algorithm
algorithm relies
relies onon information
information gathered
gathered by
by
an
an infrared
infrared sensor/receiver
sensor/receiver module
module that
that detects
detects the
the prevailing
prevailing frequency
frequency of of vibration
vibration byby aa microprocessor
microprocessor and and drives
drives the
the servomotor
servomotor
in
in accordance
accordance with with aa lookup
lookup table
table that
that relates
relates natural
natural frequency
frequency with
with angular
angular position
position ofof the
the motor.
motor. An
An analytical
analytical model
model isis developed
developed
to
to predict
predict the
the system
system dynamics
dynamics andand the
the variation
variation in in natural
natural frequency
frequency in in terms
terms of of the
the various
various design
design parameters.
parameters. While
While the
the proposed
proposed
design
design isis applicable
applicable to to various
various types
types ofof vibration
vibration energy
energy harvesters,
harvesters, anan electromagnetic
electromagnetic prototype
prototype is
is presented
presented to to verify
verify the
the concept.
concept.
Findings
Findings of of the
the theoretical
theoretical model
model areare verified
verified experimentally.
experimentally.
© 2017
© 2017
2017 TheThe Authors.
The Authors. Published
Authors. Published
Published byby Elsevier
by Elsevier
Elsevier Ltd.Ltd.
Ltd.
©
Peer-review
Peer-review under
under responsibility
responsibility of
of the
the organizing
organizing committee
committee ofof EURODYN
of EURODYN
EURODYN 2017. 2017.
2017.
Peer-review under responsibility of the organizing committee
Keywords:
Keywords: Variable
Variable inertia
inertia oscillator,
oscillator, vibration
vibration energy
energy harvesting,
harvesting, frequency
frequency tuning
tuning

*
* Corresponding
Corresponding author.
author. Tel.:
Tel.: +20226153082.
+20226153082.
E-mail
E-mail address:
address: mharafa@aucegypt.edu
mharafa@aucegypt.edu

1877-7058
1877-7058 ©© 2017
2017 The
The Authors.
Authors. Published
Published by
by Elsevier
Elsevier Ltd.
Ltd.
Peer-review
Peer-review under
under responsibility
responsibility of
of the
the organizing
organizing committee
committee of
of EURODYN
EURODYN 2017.
2017.

1877-7058 © 2017 The Authors. Published by Elsevier Ltd.

Peer-review under responsibility of the organizing committee of EURODYN 2017.
10.1016/j.proeng.2017.09.488
 Peter Ibrahim et al. / Procedia Engineering 199 (2017) 3422–3427 3423
2 Ibrahim et al./ Procedia Engineering 00 (2017) 000–000

1. Introduction

Inertial vibration energy harvesters of the linear type are usually designed to operate at resonance, hence any
changes in the operating frequency lead to a significant reduction in the output power, unless corrective action is taken
to tune the devices back to resonance. For this reason, devices possessing adjustable natural frequencies, among others,
have received attention in the research community in recent years. In these systems, the effective mass, stiffness or
damping parameters are changed to allow the system to respond to variable-frequency excitation schemes. The main
approaches adopted in the literature rely either on applying some mechanical straining forces to change the stiffness,
or on adjusting the mass distribution to change the effective mass characteristics. As tuning usually requires additional
power, current research efforts continue to address self-sufficient devices, which produce more power than they
consume, and self-adaptive devices, which autonomously adjust their frequencies without manual intervention. For a
comprehensive review of frequency tuning methods for piezoelectric energy harvesting systems, the reader is referred
to the work of Ibrahim and Ali [1]. Other useful reviews on techniques to increase the operating bandwidth of energy
harvesters are presented by Zhu et al. [2] and Twiefel and Westermann [3].

Efforts to alter the natural frequency through the application of straining forces include the work of Eichhorn et al.
[4] in which axial preload was applied by a piezoelectric actuator on a cantilever beam to change its resonance
frequency from 150 Hz to 190 Hz. The use of permanent magnets to apply forces on structures has received wider
attention because of the ease of applying variable magnetic forces by adjusting the separation between the magnets in
action. In this context, reference is made to the work of Aboulfotoh et al. [5], which presented a tunable harvester with
magnetic attraction forces to automatically adjust the natural frequency of a cantilever beam in the range from 4.7 Hz
to 9.0 Hz. Magnetic forces, both attractive and repulsive, were applied by Challa et al. [6] to alter the stiffness of a
piezoelectric cantilever beam. More recently, Hoffmann et al. [7] presented a self-adaptive electromagnetic cantilever
beam which utilizes a rotatable magnet to tune the frequency from 30 Hz to 50 Hz via magnetic forces, whereas Dong
et al. [8] proposed a tunable energy harvester where the frequency can be tuned via the application of magnetic force
in two dimensions to alter the transverse and axial stiffness of a piezoelectric cantilever beam. Efforts to change the
geometry of a system to vary the stiffness have recently received attention. In this context, reference is made to the
work of Heit et al. [9] in which a vibration energy harvester was designed consisting of a wishbone spring, whose
geometry can be adjusted, thereby changing the structural stiffness. Reference is also made to the work of Lee and
Chung [10] in which a rotatable spring with a variable stiffness was developed.

While extensive efforts have been expended on adjusting the stiffness characteristics of structures through the
application of straining forces, less emphasis was granted on adjusting the mass properties. System with adjustable
mass characteristics are likely to consume less power in tuning, since no mechanical straining forces are to be exerted
and maintained, but rather a mass redistribution is needed to cause a change in natural frequency. In this work, we
address this topic by proposing to adjust the rotary inertia of a cantilever-type energy harvester in order to enable the
device to exhibit a variable resonance frequency, thereby enabling it to operate across a variable frequency range.

2. Design

Figure 1 shows a schematic illustration of the proposed variable inertia energy harvester. The concept relies on a
cantilever beam with a tip mass carrying a pair of tuning masses that are connected by an arm. The beam undergoes
transverse vibrations (i.e. the cross section rotates about the Y-axis). Changing the angular position  of the tuning arm
(about the Z-axis) results in a change in the effective mass moment of inertia about the Y-axis of the cantilever beam.
In this way, the natural frequency of the system can be adjusted. The limiting cases are when  = 0 (maximum moment
of inertia; minimum natural frequency) and  = 90° (minimum moment of inertia; maximum natural frequency.) To
realize this concept, the mass is driven by a beam-mounted servomotor which controls its angular position about an
axis normal to the beam’s mid-plane. Energy can be harvested by electromagnetic induction, though the concept can
equally be applied to piezoelectric devices. The servomotor is equipped with a gearbox to prevent it from being back-
driven by the tuning arm.
3424 Ibrahim
Peter et al./ et
Ibrahim Procedia Engineering
al. / Procedia 00 (2017)
Engineering 199000–000
(2017) 3422–3427 3

Y
Tuning mass

Tuning arm Pickup coil

X Cantilever beam
Z 

Magnet
Servomotor

Figure 1. Proposed concept of variable inertia energy harvester.

3. Dynamic Model

The main system parameters are illustrated in Fig. 2(a). As the system operates close to its fundamental resonance
frequency, an approximate single-degree-of-freedom model, as depicted in Fig. 2(b), is developed to study the dynamic
behavior. The model consists of an equivalent mass m, attached by a spring k and a damper c to a base undergoing a
displacement y. The displacement is denoted by x and electromagnetic induction causes an electric current i to flow in
a circuit that is assumed to have a resistive load RL.

i RL
lta mtm

mta x
mm m
mtm mg
c k
X y
L

(a) (b)
Figure 2. (a) Variable inertia oscillator; (b) equivalent single degree of freedom model.

The effective stiffness of the cantilever beam can be expressed as:

k  3EI L3 (1)
where E is Young’s modulus of elasticity of the beam, I is the second moment of area of beam, and L is the beam
length. Rotary inertia effects of the cantilever beam itself are ignored as they are negligible compared to the inertia of
the tuning arm and tuning masses. Accordingly the effective mass m, which includes the effect of the tip mass mt, the
mass moment of inertia of the proof mass It, and the beam mass mb, can be expressed as [11]:
m mt  0.24mb  3I t L2 (2)
where, the tip mass mt is the combined mass attached to the beam’s tip, which is given by:
mt  2mtm  mta  mm  mg (3)
where mtm is the mass of each tuning mass, mta is the mass of the tuning arm, mm is the motor mass and mg is the
magnet mass. In equation (2), It is the total mass moment of inertia of the tip masses about the out-of-plane axis, and
is given by:
 It 121 mta lta2 cos2 ( )  12 mtmlta2 cos2 ( ) (4)
where lta is the length of tuning arm. The first term in Equation (4) accounts for the mass moment of inertia of the
tuning arm, and the second term accounts for the tuning masses, and both terms depend on the angle  of the tuning
arm. The electromechanical model follows the formulation by Poulin et al. [12] and the governing equations of motion
of the system can be expressed by:
mx  cx  kx  ky  cy  Bli (5)
 Peter Ibrahim et al. / Procedia Engineering 199 (2017) 3422–3427 3425
4 Ibrahim et al./ Procedia Engineering 00 (2017) 000–000

and
U Bl dx dt  Rei  Le di dt (6)
where U is the voltage across coil terminals, B is the magnetic field, l is total length of wire in the coil, Re is the coil
resistance Le is the coil inductance, and t is the time. To verify the concept, a steel cantilever beam carrying a tuning
arm and tuning masses was constructed. The beam was disturbed from equilibrium and the resulting free vibration
was monitored by a non-contact proximity probe. The frequency of vibration was determined at different angles of
the tuning arm. The selected dimensions and material properties are listed in Table 1.

Table 1. Dimensions and material properties.

Parameter Value Parameter Value

L 140 mm lta 250 mm
b 28 mm mta 12 g
h 0.7 mm mtm 5g
E 210 GPa mm 10 g
 7850 kg/m3 mg 3g

Figure 3 shows a comparison of the theoretical predictions of the natural frequency and the experimental findings.
The natural frequency increases with the tuning arm angle. Increasing the angle beyond 90 degrees is not necessary
due to symmetry. For the parameters specified, the natural frequency can be tuned from 8 Hz to 10.3 Hz.
10.5
Model
10 Experiment
Natural Frequency [Hz]

9.5

8.5

7.5
0 20 40 60 80 100
Angle of Tuning Arm [Degrees]

Figure 3. Variation of the natural frequency with angle of the tuning arm.

4. Frequency Tuning Algorithm

Once a relation is established between the natural frequency and angle of the tuning arm, a frequency tuning
algorithm is developed to keep the beam operating at resonance. The algorithm relies on information gathered by an
infrared sensor/receiver module that detects the prevailing frequency of vibration by a microprocessor. The sensor is
mounted on a fixed frame and works by counting the number of times the beam tip interrupts the receiver module as
the beam vibrates. A pulse is generated every time an interruption takes place. The frequency is then calculated in the
microprocessor by counting the number of pulses in a specified time interval (process time), which was taken to be 5
seconds in order to capture several interruption cycles for a better accuracy. By comparing the incoming frequency
(from the sensor) and the natural frequency (through the current angular position of the motor), the system works to
adjust the angle of the tuning arm by driving the servomotor. Once activated, the system works in a continuous loop:
sensing the base excitation frequency, comparing it to the current natural frequency (set by current angle) and driving
the servomotor to achieve resonance. A DC servomotor with a torque of 1.5 kgcm is used, which is controlled by a
pulse signal that is provided by an Arduino Uno micro-controller. A photo interrupter sensor (type OS25B10)
combining an infrared light emitting diode and a sensitive photo transistor is used. As the beam passes between the
diode and transistor a signal is sent to the micro-controller to calculate the frequency of vibration. A slotted cardboard
sheet is attached to the cantilever tip to interrupt the path between the sender and receiver points of the sensor. Data
sent by the sensor is processed in the micro-controller to capture the excitation frequency with a resolution of 0.1 Hz.
3426 Peter Ibrahim et al. / Procedia Engineering 199 (2017) 3422–3427
Ibrahim et al./ Procedia Engineering 00 (2017) 000–000 5

5. Experimental Work

Figure 4 shows a photogtaph of the experimental setup. A cantilever beam is mounted on a shaker (Brüel &
Kjær type 3386-062). The tuning masses are connected by a steel rod, which is mounted on the servomotor shaft. The
servomotor is attached at the beam’s tip. The beam vibrates in the horizontal plane to avoid any sagging due to
gravity.The input signal is generated by a function generator (HP type 3314A) and fed to a power amplifier (Brüel &
Kjær type 2706). The output voltage generated by the coil is read directly on an oscilloscope, hence the output voltage
readings are close to the open-circuit condition. The beam is excited harmonically and the frequency is set by hand.

Figure 4. Experimental setup.

Figure 5 shows the experimental and numerical output voltage in the frequency domain for three different values
of the tuning arm angle. The amplitude of the shaker displacement was fairly constant across frequency and was kept
at 1 mm. The magnetic field was estimated at 1 mT and is assumed to be constant in the numerical model. A damping
ratio of 0.017 was assumed for the beam. A coil having an outer diameter of 17 mm, inner diameter of 11 mm and a
height of 20 mm was used in this work. The coil was found to have a resistance of 2.59 k. The length of wire making
up the coil was estimated to be 50 m and the coil inductance was estimated to be 1 mH. Inspection of the plots reveal
an increase in the output voltage with angle, which is attributed to the increased resonance frequency. Discrepancies
in the voltage magnitudes can be attributed to different damping characteristics at different tuning arm angles.

100 100
0 Degrees 0 Degrees
45 Degrees 45 Degrees
80 80
90 Degrees 90 Degrees
Output Voltage [mV]

Output Voltage [mV]

60 60

40 40

20 20

0 0
5 6 7 8 9 10 11 12 5 6 7 8 9 10 11 12
Frequency [Hz] Frequency [Hz]
(a) (b)
Figure 5. (a) Experimental and (b) numerical output voltage for three different angles of the tuning arm.

Figure 6 illustrates the autonomous operation of the system and its capability of self-adjusting its natural frequency to
match that of the incoming vibration. For the scenario shown, the beam is first excited at 8.8 Hz, whereas the tuning
arm is initially positioned at an angle of zero, which corresponds to 8 Hz. It takes the system 5 seconds to sense the
frequency mismatch, and the servomotor abruptly turns, placing the tuning arm at an angle of 50° thereby achieving
resonance. The excitation frequency is kept constant for 16 s, and is then manually changed to 9.4 Hz. Once this
happens, the system operates off-resonance, as indicated by the reduced output voltage at approximately 21 s. It takes
the system two adjustment cycles to change the angle of the tuning arm to 64.3° to achieve resonance again at 30 s.
The excitation frequency is maintained for about 9 more seconds, and is then increased to 10 Hz. Once again, the
system responds after 5 s and successfully tunes itself to the incoming frequency by changing the arm’s angle to 78°,
which leads to an increase in the output voltage shown in the last segment of the time history plot.
 Peter Ibrahim et al. / Procedia Engineering 199 (2017) 3422–3427 3427
6 Ibrahim et al./ Procedia Engineering 00 (2017) 000–000

Figure 6. Time history plots of the input frequency, tuning arm angle and output voltage.

6. Conclusions

A variable inertia energy harvester was presented in this work. The device is capable self-adjusting its natural
frequency from 8 Hz to 10.3 Hz by varying the angular position of a rotatable tip mass mounted on a cantilever beam.
An infrared sensor/receiver module is used to detect the frequency of oscillation and a microprocessor is used to drive
a servomotor to maintain the system in a state of resonance. It is expected that this design draws little power for
frequency adjustment, since power is only needed to change the orientation of the tuning mass. The system was studied
analytically and experimentally, and was able to follow the applied forcing frequency if it was changed arbitrarily.

References

[1] S.W. Ibrahim and W. G. Ali, A review on frequency tuning methods for piezoelectric energy harvesting systems, Journal of Renewable and
Sustainable Energy 4, 062703 (2012)
[2] D. Zhu, M.J. Tudor and S.P. Beeby, Strategies for increasing the operating frequency range of vibration energy harvesters: a review,
Measurement Science and Technology, 21 (2) (2010) 022001 (29pp)
[3] J. Twiefel and H. Westermann, Survey on broadband techniques for vibration energy harvesting, Journal of Intelligent Material Systems
and Structures, 24(11) 1291–1302, 2013.
[4] C. Eichhorn, R. Tchagsim, N. Wilhelm and P. Woias, A smart and self-sufficient frequency tunable vibration energy harvester, Journal of
Micromechanics and Microengineering, 21 (10) (2011) 104003 (11pp).
[5] N.A. Aboulfotoh, M.H. Arafa, and S.M. Megahed, A Self-Tuning Resonator for Vibration Energy Harvesting, Sensors and Actuators A:
Physical, 201 (2013) 328-334.
[6] V. R. Challa, M G Prasad and F. T Fisher, Towards an autonomous self-tuning vibration energy harvesting device for wireless sensor
network applications, Smart Materials and Structures 20 (2011) 025004 (11pp).
[7] D. Hoffmann, A. Willmann, T. Hehn, B. Folkmer, and Y. Manoli, A self-adaptive energy harvesting system, Smart Materials and
Structures, 25 (2016) 035013 (10pp).
[8] L Dong, M G Prasad and F.T. Fisher, Two-dimensional resonance frequency tuning approach for vibration-based energy harvesting, Smart
Materials and Structures 25 (2016) 065019 (14pp)
[9] J. Heit, D. Christensen and S. Roundy, A vibration energy harvesting structure, tunable over a wide frequency range using minimal
actuation, Proceedings of the ASME 2013 Conference on Smart Materials, Adaptive Structures and Intelligent Systems (SMASIS2013),
September 16-18, 2013, Snowbird, Utah, USA.
[10] B.-C. Lee and G.-S. Chung, A novel frequency tuning design for vibration driven electromagnetic energy harvester. IET Renewable Power
Generation 9 (7) 2015, 801 – 808.
[11] R.D. Blevins, Formulas for Dynamics, Acoustics and Vibration, John Wiley & Sons, Ltd, 2016.
[12] G. Poulin, E. Sarraute and F. Costa, Generation of electrical energy for portable devices, Comparative study of an electromagnetic and a
piezoelectric system, Sensors and Actuators A 116 (2004) 461–471.