User's Manual

STUDY OF MAGNETIC LEVITATION SYSTEM

Model: ML-01
(Rev: 01/04/2010)

Manufactured by:
Techno Instruments
261/16, Civil Lines
Roorkee -247 667 UK
Ph.:01332-272852, Fax: 274831
Email: [email protected] ISO 9001:2015
[email protected] Certified Company
Website: www.sestechno.com
 CONTENTS

Section

Copyright, Warranty, and Equipment Return Page

1. Objective

2. Background Summary

2.1 System Model

2.2 Controller Scheme 2
2.3 Parameter Determination 3
3. System Description A
3.1 Suspended Object
3.2 Electromagnet
3.3 Power Amplifier4 A
3.4 Position Sensor 4

3.5 Error detector and

controller
4.
Experiments and Results
4.1 System ldentification
5. Further Work
6. References
7. Packing List
8. Technical support 8

9. List of
Experiment
 cOPYRIGHT AND WARRANTY

Please-Feel free to duplicate this manual subject to the copyright restriction given below.

cOPYRIGHT NOTICE

The Techno Instruments Model ML-01 Study of Magnetic Levitation System manualis
copyrighted and all rights reserved. However, permission is granted to non-profit educational
institnutions for reproduction of any part of this manual provided the reproduction is used only
for their laboratories and is not sold for profit. Reproduction under any other circumstances,
without the written consent of Techno Instruments is prohibited.

LIMITED WARRANTYY

Techno Instruments warants this product to be free from defects in materials and
workmanship for a period of one year from the date of shipment to the customer. Techno
Instruments will repair or replace, at its option, any part of the product which is deemed to be
defective in material or workmanship. This warranty does not cover damage to the product
caused by abuse or improper use. Determination of whether a product failure is the result of
manufacturing defect or improper use by the customer shall be made solely by Techno
Instruments. Responsibility for the return of equipment for warranty repair belongs to the
customer. Equipment must be properly packed to prevent damage and shipped postage or
freight prepaid. (Damage caused by improper packaging of the equipment for return shipment
will not be covered by the warranty). Shipping costs for returning the equipment, after repair,
will be paid by Techno Instruments.

EQUIPMENT RETURN

Should this product have to be returned to Techno Instruments, for whatever reason, notify
Techno Instruments BEFORE returning the product. Upon notification, the return
authorization and shipping instructions will be promptly issued.

Note No equipment will be accepted for return without an authorization.

When returning equipment for repair, the units must be packed properly. Carriers will not
for damage by improper packing. To be certain the unit will not be
accept responsibility
damaged in shipment, observe the following rules:
. The carton must be strong enough for the item shipped.
2. Make certain there is at least two inches of packing material between any point on the
apparatus and the inside walls of the carton.
3. Make certain that the packing material can not displace in the box, or get compressed,
thus letting the instrument come in contact with the edge of the box.

Techno Instruments ML-01 Page

 STUDY OF MAGNETIC LEVITATION
MODEL: ML-01
888 SENSOR vOLTAGE
MAGNET CURRENT

MAGNET UNIT
POWER
cONTROLLER GAIN AMPLIFIER ELECTRO
MAGNET

oBJECT

11 SENSOR
REFERENCE

SENSOR
ON
TTECHNO NSTRUMENTS, Pame

50

50

MAONETIC LEMTAnON exPERIMENT

TECHNOo INSTRUMENTS, A a

Panel Drawing Magnetic Levitation, Model ML-01

STUDY OF MAGNETIC LEVITATION SYSTEM
ML-01
1. OBJECTIVE

Theoretical and experimental study of a magnetic levitation setup,

unstable system.
an
inherently

2. BACKGROUND SUMMARY
The magnetic levitation system consists of an
clectromagnet which pulls an object (a
magnetic material) in an upward direction, in the presence of downward gravitational force on
it (Fig.1). If the magnet current i is adjusted to satisfy the condition, f= mg, the object should,
at least theoretically, remain suspended in air. In a
practical situation, however, even the
smallest disturbance would dislocate the balance and the object would either stick to the
magnet or fall down to ground. Logically therefore the current i needs to be continuously
adjusted to keep the object freely suspended in air. This task is impossible to be achieved
manually, and therefore needs a feedback control loop.
The basic feedback control scheme is shovwn in Fig.2. The idea here is to monitor the
position of the object continuously and adjust magnet current automatically to ensure the
upward force, f, exactly balances the weight of the object, mg, at all times. It will however be
seen later that due to the unstable dynamics of the object, the automatic control scheme of
Fig.2 is not workable and hence a more elaborate controller is required. A linearized model of
the system is developed next.

2.1System Model
The electromagnet pulls the object (iron ball) with a force,

f(x,i) =
and e=a~o (1)
Here,
x is the distance between the magnet and the object
i is the current in the magnet coil
c is the constant of proportionality
Lo is the additional inductance of the magnet coil due to the object placed at the
nominal position, x Xo.
The equation of motion of the object is given by

m=-o +mg,where (2)

m is the mass of the object, and

g is the acceleration due to gravity.

A linearized form of the above equation may be derived as,

mX)-e mg 3)

Techno Instruments ML -01 Page 2

 Electromagnet

Object

mg

Fig.1 Basic system

Magnet
Cumrent Force Object
Desired Position
Power Electromagn f Object
-
position of
the object Amplifier et Dynamics
Vref

Position
Sensor
Measured position
of the objed

Fig.2 Automatic Scheme for Control

Incremental voltage applied

to the megnet coil Incremental position
Ko of the object

(s+a)s-B)

Fig.3 Open Loop System Model

 current around
where, (t) and i (1) are the incremental displacement and incremental magnet
assumed to be at rest at x X, with
=

their nominal values, xgand io. Since the object is

i l0

mg, which leadsto

initial condition,
Taking Laplace transform and neglecting
-2cig
mXo ...4)
T) s22ci,

mx
is obtained as the dynamics of the object.
The current- voltage relation of the magnet is given by

the incremental values of

vt)=R *t) + L
dt
, wherevt)Jand +(t) are

and R, L are the coil

voltage applied to and the current flowing into the electromagnet,
parameters. Taking Laplace transform and neglecting initial conditions,

vs)= (R+sL) I(6), and

the transfer function
combining the above equations, the system dynamics is given by
2clo
Ris)- G$) mLx K ..(5)
(s+a)s-B*)
mx

2.2 Controller Scheme

The above equation may be represented by the block diagram of Fig. 3 which is the

open loop system.

It is clear that the open loop system above is unstable, due to the pole at s = ß, in the

line of Fig.2 will not lead to

right half of the s-plane. Also, connecting a feedback loop on the
a stable closed loop system for any value of forward path gain as may be seen from the root
locus diagram of Fig.4. This obviously is a proportional controller, which will not stabilize
the system. In case however a proportional - derivative (PD) controller is used, the revised
root locus diagram of Fig.5 indicates the possibility of stabilizing the system and also
achieving good transient performance for some range of values of forward path gain.
The PD controller is assumed to have the transfer function, Ge=(1 +Tas)
The present cexperiment involves determination of the system parameters, Ko, a and p
experimentally, and then to design Td, the parameter of the controller and choose an

appropriate gain.

Techno Instruments ML 01 Page 3

 - a

Fig.4 Root Locus with Proportional Controller

,jo

-a -T B

Fig.5 Root Locus with PD

Controller
2.3 Parameter determination
(a) The value of a can be calculated from the magnet parameters, R and L, and using
the expression a = R/ L.

(6) Since the open loop system, eqn.(5) is unstable, no experimentation is possible
directly on it. An alternative method is to use an adhoc setting of the PD controller

to result in stability. Then the force balance equation, c mg,is used to

calculate an average value of c from measurements of i and x around the nominal

equilibrium point io, Xo. The mass of the object, m, is known beforehand.
The constant Ko and ß are then computed from

Ko 2ci and B=,

mLx Vmx,'
3. SYSTEM DESCRIPTION
The hardware involved in the various blocks of Fig.2 are described in this section.

3.1 Suspended Object

This comprises of two disc magnets Mi and M2 fixed on the two ends of a
height
plastic cylinder (Fig. 6). While Mi is used for providing upward force to pull the object, M2 is
used for generating the necessary position information through the hall sensor fixed on the
base. A cylinder made of plastic material ensures proper view and protected movement of the
object. Note that our earlier analysis refers to iron as the suspended object, in the actual model
a magnet Mi is used. This ensures a better control when the object goes close to the
electromagnet. Also since the thickness of a magnet is much smaller that Xo, the calculation
are not too much in error. Finally, since the
system is highly non-linear and we are linearizing
around the operating point xo, io, the whole analysis is reasonably accurate even for a
magnet
object.
3.2 Electromagnet
A powerful electromagnet is mounted on the top of the guide rod. It exerts variable
attraction force on M depending on the current supplied to it. The second
magnet M2, being
farther away, is assumed to have a negligible effect on the force. The electromagnet is
characterised by its winding resistance R and inductance L.

3.3 Power Amplifier

It is a complementary symmetry power amplifier having an internal gain of 2, and
capable of providing up to 3 amp. to the electromagnet. The extenal gain of the circuit may
be varied form 1 to 1l using the
potentiometer on the panel.

Techno Instruments
ML 01 Page 4
 Electromagnet
Magnet, M

M,

Transparent
Cylinder

M2
Hall Sensor
Sensor, SS

Fig.6 Mechanical Arrangement of the Object

ERROR DETECTOR
CONTROLLER
R=100K
Ve N
R,= 10K
sense

C1

Fig.7 Electronic Circuit of the

System
3.4 Position Sensor
It is a hall effect device, which
generates an electrical signal propprtional to is
distance from M2. The sensor
output is thus a measure of the position of the suspended
object. Note that due to the highly non-linear nature of the overall system, we are
concerncd with small deviations around the basically
the sensor as a
equilibriurm point. This enables us to reprev
constant gain for Computation purposes
3.5 Eror detector and controller

The circuit diagram of this section is shorwn in Fig 7. It performs the tasks of,
() Placing the object to a suitable position which may be taken as the 'omina
position. This is done by adjusting the reference signal, V
ii) Providing a PD controller to ensure system sability. and a simple caloulation
shown the controller transfer function to be, (1+sR,C,).
R
(ii Setting the forward path gain to an appropriate value.
4. EXPERIMENTS AND RESULTS
The experiments conducted on a typical unit and the results are described below in
some detailed. The student is expected to repeat the steps on the unit zvailable to him and
through the design. Actual results are likely to be somewhat different in each case.

4.1 System ldentification

Standard Parameter values of a typical unit are:

Magnet Inductance, L =8.69x 10h

Magnet Resistance, R - 4.95 0

Mass of the Object, m =

10 1 Kgg
(a) Based on the parameters of the magnet, the time constant of the coil may be
calculated as,
4.95 569.62 sec
8.69x103

(b) The steps below are now followed to determine the remaining parameters
Step.1 Connect the PD controller (C= 24F) and close the feedback loop. Set
forward gain to a medium value say 4. The object should now be suspended
freely in air by adjusting Vre.
Step.2 Adjust the reference input to move the object in the gap between the magnet
and the sensor. This is the nominal position of the object [xo is the disance
between the object (top marker) and the electromagnet. M and is is the
electromagnet current.]
Step.3 Record xa, io and calculate c. The mass of the object,

m 10 10 Kg (Specified by the manufacturer)

X = 34 mm = 0.034 m

io = 1.15 amp

Techno Intruments
 Root Locus

500

400

300

200

100

-100

-200

-300

400

-500
-800-700-600-500 400 -300 200 -100 0 100 200
Real Axiss

K.3922
Gs)+569)-24)
Fig. 8 Root Locus Diagram of the Uncompensated System

Impulse Response
0.3

0.2

0.1

-0.1

-0.2

-0.3

04
0.02 0.04 0.6 0.08 0.1 0.12 0.14 0.16 0.18 0.2
Time (sec)

1+Gs)
Fig.9 Impulse Reponse
of the uncompensated
System
 Cmg 10x10x9.8034-8.56x10
io
Step.4 Compute Ko and B(Ref to sec. 2.3) as,

Ko = 2cio 2g 2x9.8
= 1961
mLx2 Lio 8.69 x10 x1.15
2
B 2ci 2g2x9.8
V 0.034
=24.00
mxo
Step.5 The transfer funetion of the system at the nominal position is now written
explicitly as
2x 1961
Gs)= (Internal gain of power amplifier = 2)

(s+569)4-24.00)
The forward path transfer function may therefore be written as
Kx3922
K.Gs)=-G+569)s -24.00
Where K is adjustable between I and 11 as explained in section 3.3
Step.6 Feedback path gain (sensor gain) at the nominal position is found by
displacing the object slightly around the nominal position (say #3mm) and
monitoring the change in sensor output.
i) Yo 5 mm Sensor output, V, = 3.62 V

(i)yo +Ay= 8mm Sensor output, Vs +AV,=3.25 V

(iin) y,-Ay= 2mm Sensor output, V,- AV,=4.63 V

Average Sensor gain Ks= 2AV, 230 V/m

24 y,

The feedback path transfer function may therefore be written as

H(s) = 230

Step.7 Sketch the root locus diagram from the transfer function obtained in Step. 5
and 6 using MATLAB or otherwise, as shown Fig.8. Instability of the closed
loop system for all values of open loop gain is obvious, and may also be
verified through an impulse response plot as in Fig. 9.
Note that since the system has a non-unity feedback path transfer function, the root
locus plot must take this into account. Sample MATLAB commands for this
purpose
are listed below.

Techno Instruments ML -01 Page 6

 500
Gain:3.04
400 Pole-53.1+ 49.1
Damping 0.734
300 Overshoot (%): 3.34
Frequency (rad/sec): 72.3
200
100
100 F
200
-300
400
-500
400 -300 -200-100 0 100 200
-800 -700-600 -500
Real Axis
GGK(1+0.02s)x 0.3922.230
s+569)s-24)
Fig. 10 Root Locus Diagram of the Compensated System
0.4
0.35
0.3
0.25
0.2
O.15
0.1
0.05
-0.05
0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5
Time (sec)
Fig. 11 Impulse Response of the closed
system with K=3 Loop Compensated
 s=tf's)
g(3922)/(s+569)*(sA2-24A2):
h-230;
ghgh
rlocus(gh)
impulse((g)(+gh)), 2)
Step.8 Choose a suitable closed loop pole location (P) and design a PD controller. A
good choice may be a zero at S - 50. The compensator transfer function. Gc
=
(1+0.02s) 0.02(s+50). With Ri= 10KQ (internal), the
=

compensator capacitance is Ci 2uF. A revised root locus diagram may be

value of
drawn as shown in Fig. 10. Note that the system is now stable for K>1.5.
Impulse response plot for K - 3 is shown in Fig. 11

MATLAB command:
gc-(1/50)*(s+50))
rlocus(gh gc)
implus(3°g'gcM(1+3*gh*gc), .0:.0001:.5)
Calculate and connect the designed value of capacitor on the panel and operate
the system with the computed value of the forward path gain.
A careful observation of the root locus of Fig.10 would show that the system
could be stabilized by using a wide range of capacitors. This may be verified
experimentally.
Also it might appear that large values of K would be preferred in all cases.
This however is not true in the practical situation due to the saturation of the
amplifier.

5. FURTHER WORK
The experimental unit provides the user with an interesting platform to conduct
further experimental work. This however is most conveniently done with the support of
MATLAB. Some suggestions are:
(a) Simulate different PD controllers on MATLAB check the root locus diagram
impulse response their effect on the unit.
(b) Attempt lead compensator design and implement it.
Note: Do not expect the experimental performance to match exactly with theoretical
prediction. While the theoretical work is validfor linear system only, the experimental system
is a mon-linear one, which has been approximated as a linear system for small variation
around the operating point.

6. REFERENCES
[1 Franklin GF, JD Powell and Michael Workman, "Digital Control of Dynamic
System", Addison Wesley, 2000.
[21 Shiao Ys, "Design and Implementation ofa controller for a Magnetic Leviation
System", Proc. Natl. Sci.Comc. ROC(D), Vol. I, No. 2, 2001, pp. 88-94.

Techno Instruments ML 01 Page 7

 PACKING LIST
(1). Magnetic Levitation(Main Unit): One
(2). Patch Cords: Two
a) 40/36-One Side Crocodile Clip (6"): Two
3). Capacitor: Two
a) 1.0 uF:One
b) 2.0 uF One
(4). Magnctic Levitation Sub Unit:One

Techno Instruments ML 01
Page 8
 TECHNICAL SUPPORT

Feed Back

If you have any comments or suggestions about this product or this manual please let
us know. Techno Instruments appreciates any customcr feedback. Your input helps us
evaluate and improve our product.

To reach Techno Instruments

*Phone: +91-1332-272852, 2771 18

Fax: +91-1332 -277118
e-mail: [email protected]; [email protected]

Contacting for Technical Support

Before you call the Techno Instruments Technical Support staff it would be helpful to
prepare the following information:

I f your problem is with the Techno Instruments apparatus, note :

Model number and S. No (usually listed on the label at the
o
backsid
o Approximate age of the apparatus.
o A detailed description of the problem/ sequences of events may please be sent by
email or Fax.

I f your problem relates to the instruction manual, note;

Model number and Revision (listed by month and year on the front cover).
Have the manual at hand to discuss your questions.

Techno Instruments
ML 01 Page 9
 List of Control Laboratory Experiments manufactured by us

Digital Control System

Study of microprocessor control ofa simulated linear system

A.C./D.C. Servomotorstudy
Study of characteristics of a 2-phase a.c. motor/ d.c motor. i allows the determination of torque-speed
characteristics, inertia and friction parameters of an a.c. motor. Transfer function can also be evaluated in
this system A digital display is available for time constant measurement.
A.C.ID.C. Position Control
Study ofan a.c. /dc servomotor angular position control system
D.C. Speed Control
Studyof a dc. motor speed control system
Temperature Controller System
To study a ypical temperature control system (Compact Oven)
PID Controller
Perfomance evaluation and design of PID Controller

Study of Synchro Devices

Stuay of synchro transmitter-receiver pair with calibrated dials. Receiver can be used as control
transformer. Buil-in balanced demodulator circuit. Digital display of ac/ dc voltages.
.Linear Variable Differential Transformer
Strudy ofthe performance characteristics ofa LVDT.
Magnetic Levitation System
Analysis and design of feedback control system to keep an object suspended in air.
Stepper Motor Study
To study the operation and characteristics of a stepper motor with an 8085 based uP-Kit and user sofware
EPROM.
Relay Control System
To analyze a simulated relay control systems.
Compensation Design
Todesign a suitable cascade compensator for the given system and verify the
resulting improvement.
Study of Second Order Networks
Study of synchro transmiler-receiver pair with calibrated dials. Receiver can be
used as control
transformer. Buil-in balanced demodulator circuit. Digital display of ac/ dc
voltages.
Linear System Simulator
To study the
performance of First, Second and Third order Systems.
Potentiometric Error Detector
To study the performance a potentiometer ype d.c.
position error detector.
Light Intensity Control
P and Pl conirol
of light intensity with provision for disturbance and
transient studies
Microprocessor Device Controller
Study of 8085 P based switching control of LED,
the study
of switch state input through 8255 port andRelay sequence, 7-Segment displagy. Besides it also allow
SID/ SOD operation of a
Study of Temperature Transducers microprocessor.
Study of inpul-outpu characteristics of some common
transducers like, thermistors (PTC and
thermocouple, semiconductor sensors NTC),
Stroboscope
For measurement
of shaft speed using stroboscope principle in harsh
Function Generator laboratory emvironment.
10H-2MHz; Square/Sine/ Triangular:
Amplitude 0-3V (p-p): 4 digit digital counter
Study of Digital to Analog Convertor
Detailed study of D/A schemes
4 bit
weighted resistance, R-2R discrete network
-

circuits with 8085 based and 10-bit IC based

uP-kit and interface for CRO included
Techno Instruments
ML-01 Page 10