Human Powered Flywheel Motor - A Review

advertisement
International Journal of Engineering Trends and Technology (IJETT) – Volume 8 Number 1- Feb 2014
Human Powered Flywheel Motor - A Review
K.K.Padghan1, Prof. A.K. Pitale², Prof J.P.Modak³, A.P. Narkhedkar4
1
2
3
Scholar M.E. (CAD/CAM), PRMIT&R, Badnera, India,
Professor, Dept of Mech Engg, PRMIT&R, Badnera, India,
Emeritus Professor, Dept of Mech Engg, Priyadaeshani C.O.E. Nagpur, India,
4
Scholar M.E. (CAD/CAM), PRMIT&R, Badnera, India.
Abstract -
As man-machine system model can be established, the
machine has essentially developed on the basis of general mechanical
design. Human Power Flywheel Motor is a man-machine system model.
It consist of three main units, a pedal driven flywheel motor, the
transmission between flywheel shaft and the input shaft of process
unit, and the process unit. This literature report is review of Human
power flywheel motor; the survey proved to be system has functional
and economically viable.
to sprocket centrally which further connected with smaller sprocket which
is first shaft of energy unit and transmitted its motion by chain drive.
Key Words - Human energized flywheel motor, chaff Cutter, Brick
Making Machine, Pedal operated flour Mill, Experimental model, and
optimization.
I. INTRODUCTION
During 1979-99, Dr. J.P. Modak (1982, 1994, 1997, 1998 and n.d.)
developed human power brick-making machine for manufacturing brick
making machine. This machine human power is main energy source [1].
Human power flywheel motor the principal involved for this motor is that
this machine has proved that on average ,the power produced by human
approximately 75W (0.10 hp)[Alexandrove,1981][2]. It is possible to
mechanized process needing power far too in excess of human capacity by
using intermittent without affecting end product.
The mounting energy crisis and unemployment in
the developing countries suggest towards more effort in utilization of
human muscular power. At the time when the “Appropriateness” of
technology is being questioned daily, the bicycle which is perhaps the
most “Appropriate” and efficient machine ever intended is making a
sudden come back in area of power known as Human Power. Essentially
this machine consists of three sub system: 1) The Peddling Unit, 2)
Transmission unit, and 3) The process unit. The energy unit consists of
conventional bicycle mechanism, the transmission unit consists of clutch
and gears, process units consist of an auger, a cone, and die. [2]
II. DESCRIPTION OF FLYWHEEL MOTOR
As Alexandrove 1981 start to powered human energy to machine, with
effect human energy is energies by using intermittent like as gear and
flywheel motor. And Dr. J.P.Madak 1977 and his associate are working on
flywheel motor, a manually driven brick making machine was first kind in
which flywheel motor is used first time [5]. Essentially machine consist of
three sub system: 1) The Peddling Unit, 2) Transmission unit, and 3) The
process unit. The peddling unit consist conventional bicycle mechanism
on rigid bicycle frame. A seat is provided in the bicycle frame for sitting
purpose. This seat is adjusted on the extended portion of rod by nut and
bolt arrangement. Also this gives the comfort condition to the rider. This
unit comprises of sprocket, a pedal crank and crank set. The sprocket has
‘n’ no. of teeth as per required power transmission. Pedal crank is attach
ISSN: 2231-5381
As shown in figure1 the mechanism ‘M’ is energized by the rider by
pedaling at big sprocket chain drive ‘BSC’ there by converting the
oscillatory motion of thighs into rotational motion of counter shaft ‘Cs’.
The pair of speed increasing gears connects the counter shaft ‘Cs’ with the
flywheel shaft ‘Fs’ [4].Driver pumps the energy in flywheel at energy rate
convenient to him [4]. In this way, the muscular energy of human is
converted into kinetic energy and stored into flywheel by this man
machine system and for its efficient use it is necessary to optimize its
parameters [4].
III. PERFROMANCE PARAMETERS
Initially, the flywheel motor was developed only on the institution of
human, not based on any design data; rather it was built [5]. But later with
the numerous experimentation the design data is made available which is
discussed below.
1) Flywheel size and Moment of Inertia:
In the mechanism
having flywheel meshed arrangement, the size of the flywheel and its
moment of inertia also plays the vital role in the terminal velocity of
flywheel. Modak J.P (1987) during the experimentation has observed that
for the average person of 165 cm stature from age group 20-24 years
maximum thigh oscillation is 40. [4]. Therefore with the available chain
drive for existing 22” bicycle frame the flywheel speed of 240 rpm was
found appropriate enough from point of total speed rise from pedals to
flywheel shaft [4]. In order to store the maximum energy in flywheel
irrespective of speed fluctuation, Modak J.P.(1987) has determined the
size of flywheel (180-240 rpm)[4]. During his research the Flywheel rim
diameter is found to 82 cm which gives the weight of flywheel as 150Kg
and 266 Kg for 240 rpm and 180 rpm respectively. Hence Modak
J.P.(1987) suggested the flywheel with 150 Kg @240 rpm[4]. Further
Modak J.P. (1987) has also found that Moment of inertia has no effects on
http://www.ijettjournal.org
Page 5
International Journal of Engineering Trends and Technology (IJETT) – Volume 8 Number 1- Feb 2014
required driving torque at pedal and stores same energy for same
frequency of thigh oscillation [4].
2) Gear Ratio: The rider should feel easy to drive mechanism as well his
muscle power should be converted into maximum possible motion. In
order to provide the ease to operator it is necessary to reduce the harness
effect of vibration and jerk at the process unit shaft. Hence Modak J.P.
(1987) suggested the value of gear ratio as 4:1 so as to reduce the effect of
jerk induced at process unit shaft as result of energy or momentum
exchange during the clutch engagement. If lower value of gear ratio is to
be used then flywheel speed should be maintained higher than 240 rpm
[4].
IV. IMPROVEMENT IN EXISTING BICYCLE MECHANISM
It is modified form of mechanism called as Quick Return Ratio
One. In the existing mechanism, the ratio of forward travel to return travel
is 0.82. In the Quick Return Ratio One, the ratio is one therefore, the
second paddle will be immediately ready when the first one goes down.
In this, the thigh oscillation angle, thigh length and the leg
length are kept same. In existing mechanism, the crank length is 18.5 cm
and in QRR- one it is 20 cm.
Similarly, in existing mechanism the frame length i.e. crank centre to
rider’s hip joint 74 cm and frame inclination to vertical is 200. But in
QRR-one, the frame length ie. Crank centre to rider’s hip joint 67cm and
frame inclination to vertical is 110
2) Elliptical Sprocket :
For the efficient use of the system it is necessary to make maximum use
of the human energy. But Modak J.P. has observed that among the 360°
revolution of pedal only part of it produces the necessary useful torque.
This is because of the limitation of existing bicycle drive .Modak J.P
(1985) has established the relationship between the useful torque
developed at the crank as function of crank position during its revolution
[8]. The observations made by Modak J.P. are tabulated below.
Serial no.
1
Crank position from TDC
0°-30°
Torque
Partially useful
2
30°-115°
Useful
3
115°-162
Partially useful
4
162°-360°
Idle
Table 1: Relation between crank position and torque produced.
Even when both the cranks are considered the useful driving angle is
found to be 154°. [8].Consequently for improvement in maximum
utilization of operators energy Modak J.P. suggested three modified
mechanisms namely Quick return ratio one, Double lever inversion and
Elliptical sprocket[8].Based on his mathematical modeling he concluded
improvement of 18%,38% and17%, in human energy utilization for
Elliptical sprocket, Quick return ratio one, Double lever inversion and
respectively. This performance of various bicycle drives then was
experimentally verified by Modak J.P., Chandurkar K.C. (1987) and
found almost matching with theoretical values [7].
1) Quick return ratio-one :
Fig 3. Modified Mechanisms (Elliptical Sprocket)
In modified mechanism, an elliptical sprocket is at the crank and circular
sprocket at the wheel. The main objective of using elliptical sprocket to
such a drive can be that the chain becomes alternately loose and tight.
From figure, an elliptical sprocket, chosen, the chain becomes
loose by 0.8 cm only from its tight condition and sag of only 3 cm is
produced by its weight in the span of 49.5 cm. Here an elliptical sprocket
of major diameter 24 cm and minor diameter is 12 cm. chosen so as to
keep the chain length same.
Assuming that the chain speed is constant, the elliptical
sprocket will rotate at a higher speed when the transmission angle is poor
and will rotate slow when the transmission angle is good. Thus, giving
more time to the mechanism for receiving power input.
3) Double Lever Inversion
Fig2: Modified Mechanism (Quick Return Ratio = 1)
From figure
Fig 4. Modified Mechanisms (Double Lever Inversion)




From figure, O2A is a crank which does not rotate completely but
oscillates. Therefore, O2A is chosen to give sufficient angular
displacement and good transmission angle. The lever centre or crank
centre O2 was located on the perpendicular bisector of A1A8 so as to give
an oscillation angle A1O2 A8 of 700.
O1B is thigh length,
AB is length,
O2 A is crank length,
O1 O2 is frame.
ISSN: 2231-5381
http://www.ijettjournal.org
Page 6
International Journal of Engineering Trends and Technology (IJETT) – Volume 8 Number 1- Feb 2014
In double lever mechanism, lever O2 A is 32 cm long and frame
O1 O2 is 58cm at an angle 100 to vertical. Another four bar chain O2 CDO3 is
provided in series. This auxiliary four bar chain is a crank lever inversion
with crank O3 D rotating and lever O2C oscillating along with O2 A. In
addition to timing the lever, oscillates in positions O2A1 to O2 A8 the
auxiliary mechanism serves an additional purpose, when lever for one leg
is moving down, it moves the lever of another leg upward. The crank O3 D
for both the levers is kept 1800 out of phase. The auxiliary mechanism is
designed for a quick return ratio less than one to ensure that the two levers
of different height do not come to dead centre position at the same time.
V. PLANNING OF EXPERIMENTATION
The dimensional equation for the process is deduced as follows:
such as force exerted on pedal/crank, measuring crank angle w.r.t. frame,
measuring angle between pedal w.r.t. ground, Human input energy
measurement (R). Modak J.P. and Bapat A.R. (1994) formulated
generalized experimental model for flywheel motor [3]. They established
the functional relation between the terminal angular velocity (W) and
other dependent variable such as moment inertia (I), Gear ratio (G),
Human input energy (R), Effectiveness of mechanism (EM). From this
functional relation, for a particular time period of oscillation (T), the
terminal velocity at the end of pedaling can be determined. [8] Further
during experimentation Modak J.P. and Bapat A.R. (1994) also find the
variation of terminal velocity (W) with G, I, and EM. Determined the
valued of G, I, and EM depending upon the objective of study. Based on
the deduced model, the optimum values for the independent parameters
for various objective functions can be produced as described in table 3. [3]
WT = f [(I/RT2), (ME), (G)]
Where,
W = Angular Velocity of flywheel in rad/sec reached after time interval T
secs.,
I = moment of inertia of flywheel, Kg-m2
R = energy input by rider, Kgf-m
ME = effectiveness of mechanism, ‘M’
G = speed increasing gear ratio 16
f = stands for function of
T = peddling time, in second.
Table 2 enumerates Test envelope, Test points and Test sequence [11]
for every independent P term or parameter which have been worked out
Based on earlier findings [1 and 9-10]. It is not possible to estimate
Test envelope for (I/RT2) because it is not possible to adjust R for the
Given rider. Hence Test envelope for parameter I only am considered.
Sr.
No.
1.
pi Terms
Test Envelope
Test Sequence
‘I’ Moment of Inertia
of Flywheel, kg-m2
0.255 to
(for I)
0.255,1.867,
3.48,
1.061, 2.673
2.
‘ME’ Effectiveness
of
Mechanism “M”
1 to 1.18
Gear Ratio
1.14 to 4.0
3.
Eм
G
I (kg m²)
Maximum WT
1
4
0.255
Maximum energy storage
1
4
0.255
Maximum effectiveness
1
2-4
0.255-1.061
Optimized
obtaining..
3.48
for
Table 3: Optimized Value
Modak J.P. and Bapat A.R. [1] conducted experiments for various
combination of moment of inertia (I) & (G) and determined the variation
of pedal force (Ft) with position of crank angle. Modak J.P. and Bapat
A.R. [1] also found that of moment of inertia (I) of flywheel should not
between 1.4 to 2.4 Kg/m2 because during this variation pedal force (Ft)
and load torque to overcome has a maximum value. Furthermore in order
to minimize frictional losses the gear ratio (G) should be taken 3.8 and
moment of inertia (I) should 1.06 Kg/m2.
VII. CONCLUSION
1, 1.17, 1.18
1.14, 1.5,
2.0, 1.3
4.0,
Table 2 : Test Envelope, Test Points and Test Sequence
The different bicycle mechanisms developed by Modak [10] other than
existing bicycle mechanism are i) Quick Return Ratio = 1 drive and ii)
Double lever mechanism. The Elliptical drive mechanism [14] is also
studied. It is found theoretically that effectiveness of Q.R.R.=1 drive,
Elliptical drive and Double lever inversion is respectively 17%, 18% and
35% more effective as compared to existing drive from the point of view
of rider’s energy utilization [18]. Hence in TABLE 2 the test envelope is
Taken as 1 to 1.18 for ME in view of the fact that in the present
Investigation only three mechanisms viz. (1) Existing (2) Q.R.R. =1 drive
And (3) Elliptical sprocket drive [19] are tried. Double lever inversion,
Although should have been tried in spite of the fact that it’s ME value is
1.34 could not be tried at this stage because of it’s mechanical fabrication
Complexity. The physical design and fabrication of the experimental set
up was then carried out [6 & 7] and the set up was tested for sturdiness,
accuracy and smooth running [3]. A procedure is evolved [10] for
eliminating the effect of extraneous variables associated with rider like
temperament, attitude, day and duration of test etc. on the experiment.
VI. OPTIMISATION OF VARIOUS PARAMETERS
As the flywheel motor has wide range of applications, it is being constant
subject of researchers for the performance improvement of the system
through parameter optimization. For this various experimentation is done
on the flywheel motor. To determine the various dynamic responses
,Modak J.P. and Bapat A.R.( determine the various dynamic response
ISSN: 2231-5381
values
Thus the literature survey on flywheel motor is carried out. Initially the
human powered flywheel motor was developed for the manufacturing of
lime fly ash bricks. later on various applications are developed such as
chaff cutter, forge hammers, potter’s wheel etc. Because of its numerous
advantages, the flywheel motor is finding the importance in the rural side
of developing countries like India. And hence it is necessary to optimize
its performance parameters. In an attempt lots of experimental and
numerical models are developed which are already discussed. The effect
of multiple operators with alteration in the mechanisms such as double
lever inversion, quick return ratio=1, and elliptical sprocket can also be
analyzed as future work for the human powered flywheel motor.
REFERENCES
[1]
Modak J.P, Bapat A.R. “Various Efficiencies of Human
powered flywheel motor” Human power number 54: pp21-23
[2]
Modak J.P, Moghe S.D. “Design and development of a
Human-Powered machine for the manufacture of Lime-FlashSand Bricks.” Human Power Volume 13 number 2, spring
1998.
[3]
Modak J.P, Bapat A.R. “formulation of generalized
experimental model for a manually driven flywheel motor and
its optimization.” Applied ergonomics 1994 Volume 25
Number 2.
[4]
Modak J.P, “ General consideration of mechanical design for
manually driven process machine” NACOMM; 1987; pp 1317.
http://www.ijettjournal.org
Page 7
International Journal of Engineering Trends and Technology (IJETT) – Volume 8 Number 1- Feb 2014
[5]
Modak J.P, Bapat A.R “ Improvement in experimental setup
for establishing generalized experimental model of various
dynamic responses for manually energized flywheel motor”
[6]
Modak J.P “design and development of human energized
chaff cutter”
[7]
Modak J.P. , Chandurkar K.C., Singh M.P, Yadpanawar A.G
“Experimental verification of various bicycle drive mechanism
part1” Proceedings of AMSE conference modeling and
simulation Karisurhe west Germeny, july 20-22 1987;pp139160.
[8]
Modak J.P “Bicycle and its kinematics and modifications”.
National conference mach Mech; February 1985; pp5-11.
[9]
Modak J.P, Bapat A.R. “Formulation of generalized
experimental model for manually driven flywheel motor and
its optimization” Applied ergonomics; 1994; volume 25;
number2; pp 119-122.
[10]
J. P. Modak, "Bicycle-it's Kinematics and Modifications"
Proceedings of National Conference on Machines &
Mechanisms, Feb. 1984, pp 5-12.
[11]
Schenk H. Jr., “Theories of Engineering Experimentation”,
McGraw Hill Book.Co., New York, 1961.
K Inoue, K Shimada “Solid model reconstruction of
wireframe cad models based on topological embeddings of
planar graphs” - Journal of Mechanical Design, ISSN:13506501 Vol:223, Pages:1165-1177,June2003.pp273-279.
[12]
[13]
David Gordon Wilson “ Understanding the pedal power”
published by VITA 1600 Wilson boulevard; ISBN 0-86619268-9[C];1986.
[14]
K. V. Chandurkar, et all "Experimental Verification of
Various Bicycle DriveMechanisms Part-i". Proceedings of
International AMSE Conf., Modelling &Simulation Control,
KARLSRUHE (WEST Germany), vol 3-A, July 1987, pp
139-160.
[15]
A. Tyagrajan, et all "Development of on Load Starting
Positive Clutch forFrequent on-off' Proceedings of 3rd
IFToMM Speciality Symposium on RotorDynamics, Vibration
Control, France. Sept. 1990. Accepted for Presentation.
[16]
R Bogusch, B Lohmann “Computer-aided process modeling
with ModKit from northwestern.edu”- Computers & Chemical,
2001.pp173-179
[17]
J Gómez, C Cachero “Conceptual modeling of deviceindependent web applications” - Multimedia, Volume: 8, Issue:
2 IEEE, ISSN: 0278-0046 Jun 2001. pp26 – 3
[18]
Murell K. F.H., “Ergonomics, Man in His Working
Environment”, Chapman andHall Ltd., London, 1965.
[19]
William A. Williams, “Mechanical Power Transmission”,
Manual, Book Division,Conovex-Mast Publications, Inc.,
1953, p 358.
BIOGRPHIES.
K.K.PADGHAN is M.E. scholar at Prof.
Ram Meghe Institute of technology and research Badnera,
Maharashtra, India. He is basically Mechanical Engineer from
Ram Meghe Institute of technology and research Badnera,
Maharashtra, India. Presently he is perusing his post graduation
studies from . Ram Meghe Institute of technology and research
Badnera, Maharashtra, India. His area of interest is CAD/CAM,
Production engineering.
Dr. J.P.MODAK is working as Emeritus
Professor at Dept. of Mechanical Engineering, Priyadershani
C.O.E. Nagpur,India.
A.P.NARKHEDKAR is M.E. scholar at
Prof. Ram Meghe Institute of technology and research
Badnera,Maharashtra, India. He is basically Mechanical Engineer
from Jawaharlal Darda Institute of Technology, Yavatmal, India.
Presently he is peursing his post graduation studies from . Ram
Meghe
Institute
of
technology
A.K.PITALE is working as professor at .
Ram Meghe Institute of technology and research
Badnera,Maharashtra, India. He is basically Mechanical engineer
having its post graduation in mechanical engineering. His subject
of intrest CAD/CAM, thermal engineering, power engineering,
Fluid Mechanics
and research Badnera,Maharashtra, India. His area of intrest is CAD/CAM, Production engineering.
ISSN: 2231-5381
http://www.ijettjournal.org
Page 8
Download