Home Search Collections Journals About Contact us My IOPscience

Experiments with a dc motor

This article has been downloaded from IOPscience. Please scroll down to see the full text article.

2010 Eur. J. Phys. 31 863

(http://iopscience.iop.org/0143-0807/31/4/016)

View the table of contents for this issue, or go to the journal homepage for more

Download details:
IP Address: 137.112.34.235
The article was downloaded on 05/11/2010 at 18:05

Please note that terms and conditions apply.

IOP PUBLISHING EUROPEAN JOURNAL OF PHYSICS
Eur. J. Phys. 31 (2010) 863–870 doi:10.1088/0143-0807/31/4/016

Experiments with a dc motor

Yaakov Kraftmakher
Department of Physics, Bar-Ilan University, Ramat-Gan 52900, Israel

E-mail: krafty@mail.biu.ac.il

Received 9 March 2010, in final form 5 May 2010

Published 8 June 2010
Online at stacks.iop.org/EJP/31/863

Abstract
Experiments with an electric motor provide good opportunity to demonstrate
some basic laws of electricity and magnetism. The aim of the experiments
with a low-power dc motor is to show how the motor approaches its steady
rotation and how its torque, mechanical power and efficiency depend on the
rotation velocity. The tight relationship between the mechanical and electrical
parameters of the motor is clearly seen. The measurements are carried out with
the ScienceWorkshop data-acquisition system and the DataStudio software
from PASCO scientific. The experiments are well related to university courses
of electricity and magnetism and can be used in undergraduate laboratories or
for lecture demonstrations.
(Some figures in this article are in colour only in the electronic version)

1. Introduction

Like any electrical machine, a dc motor is a good teaching resource illustrating some
fundamental laws of electromagnetism. Many papers have been devoted to this issue, and only
some recent papers can be listed here (Gluskin 1998, Kraftmakher 2001, Yap and MacIsaac
2006, Lyna et al 2007, Ng et al 2009). We consider a simple dc motor, where the stator is a
permanent magnet, while the rotor contains a winding connected to a dc source of constant
voltage. A commutator and brushes serve for properly operating the motor (e.g. Smith 1984,
Hecht 1994, Bobrow 1996, Jewett and Serway 2008).
When a dc current flows through the rotor winding, the interaction between this current
and the permanent magnet (Lorentz’s force) causes the rotor to rotate. Due to the rotation,
an EMF is created in the winding (Faraday’s law). This EMF e opposes the applied voltage
(Lenz’s law) and is therefore called the back EMF (or counter EMF). During the acceleration
of the rotor, the back EMF gradually increases. When a dc voltage V is applied to the rotor
winding, the current i in the winding is governed by Kirchhoff’s loop rule:
V = L di/dt + iR + e, (1)
where L and R are the inductance and resistance of the winding, respectively. Under steady-
state conditions, the current in the winding is i = (V − e)/R. The current decreases with
0143-0807/10/040863+08$30.00 
c 2010 IOP Publishing Ltd Printed in the UK & the USA 863
864 Y Kraftmakher

Figure 1. The torque and mechanical power developed by a dc motor versus rotation velocity,
according to the linear model.

increasing rotation velocity. If a heavy load jams the rotor, the current in the winding may
dramatically increase and overheat the motor (Hewitt 2003, Whitaker 2009, Turner 2009).

2. Model of dc motor

The mechanical power P produced by the motor is proportional to the rotation velocity ω =
2π f (f is the rotation frequency) and the torque T developed: P = ωT. A good approximation
for a dc motor with a permanent magnet is a model assuming a linear dependence of the
torque on the rotation velocity (Smith 1984). Two basic points define this dependence: the
stall torque Ts (the torque reaches a maximum, but the shaft is not rotating) and the maximum
rotation velocity ωn when no load is applied to the shaft (figure 1). The maximum torque
is achieved when the rotation velocity approaches zero because the back EMF vanishes and
the current in the winding reaches a maximum value limited by the applied voltage and
resistance of the winding. Since the mechanical power P is proportional to both the torque
and the rotation velocity, the power is zero at both basic points. Accepting the linear model,
we have
T = Ts − ωTs /ωn , (2)
and
p = ωTs − ω2 Ts /ωn . (3)
Thus, the mechanical power, including the power of the friction losses, is a quadratic function
of the rotation velocity. A maximum of the power is achieved at ω = ωn/2 and equals ωnTs/4.
The above relations are similar to those for the output voltage and useful electric power of a
battery (T is an analogue to the output voltage, Ts to the EMF of the battery, ω to the current
and ωn to the maximum current limited by the EMF and internal resistance of the battery).
When the motor is unloaded and the friction losses can be neglected, its steady rotation
velocity is achieved when e = V. This relation defines the maximum rotation velocity for a
given applied voltage. Since the back EMF is proportional to the rotation velocity, the steady
rotation velocity of an unloaded lossless dc motor is proportional to the applied voltage.
When the motor is loaded, the rotation velocity and the back EMF unavoidably decrease, so
the current i in the winding and the input electric power Pel increase. The input electric power
equals
Pel = iV = i(e + iR) = ie + i 2 R, (4)
2
where the term i R shows the Joule losses in the rotor winding. This equation is only valid in
steady-state conditions.
Experiments with a dc motor 865

Figure 2. The experimental setup employing magnetic braking. The rotation velocity is measured
with a rotary motion sensor.

The torque developed by the motor is proportional to the current i, while the rotation
velocity to the back EMF e. Therefore, the mechanical power P produced by the motor should
be proportional to ie. Equation (4) shows that the mechanical power equals ie. This term
includes the power of friction losses. The equality P = ie can be derived directly from the
theory of the dc motor (Smith 1984). A strict relationship thus exists between the mechanical
and electrical parameters of the motor.
In experiments with a motor, an important question is how to create a controllable load to
it. A well-known approach is to use the motor for driving a dc generator and to determine the
electric power produced by the generator (e.g. Kraftmakher 2001, Ng et al 2009). It is easy to
change this power in a wide range and determine the friction losses. In the present experiments,
the motor rotates a conductor subjected to magnetic braking. The torque provided this way
is not determined directly; however, after the linear dependence of the torque on the rotation
velocity is confirmed, the torque becomes available from the measurement data.

3. Experimental setup

One of the aims of the experiments is the observation of the motor operation immediately after
a dc voltage V is applied to the rotor winding. In our case, this period lasts less than a second.
Therefore, we use the ScienceWorkshop data-acquisition system with the DataStudio software
from PASCO scientific.
The experimental setup is very simple (figure 2). The signal generator incorporated into
the ScienceWorkshop 750 Interface feeds a low-power dc motor (Maxon 2326.939-22.111-006
model). The resistance of the rotor winding is R = 13.7 . The shaft of the motor is connected
866 Y Kraftmakher

current in rotor winding (A)

rotation velocity (rad s−1)
4 V, unloaded
4 V, loaded

2 V, loaded
2 V, unloaded

4 V, loaded
unloaded

2 V, loaded

time (s) time (s)

Figure 3. Records of initial periods of the motor operations for the loaded and unloaded motor.
The dc voltages are 2 and 4 V. Due to different rotation velocities, the loads for the two cases are
not equal.

to the shaft of the PASCO rotary motion sensor measuring the rotation velocity ω. Since it
is difficult to precisely align the motor and the sensor, a soft connection is made between
them. An aluminium disc, 1.2 cm thick and 4 cm in diameter, attached to the other end of
the shaft of the sensor, is subjected to the magnetic field of two strong ceramic magnets, each
1.2 × 1.2 × 1.2 cm3 in size. The magnets rest on an iron slab placed on the platform of a
laboratory jack, so it is easy to change the distance between the magnets and the disc. With
the magnetic braking, the load on the motor can be changed in a sufficiently wide range. To
obtain maximum magnetic flux penetrating the aluminium disc, magnets are aligned as shown
in figure 2. The current in the winding i is taken as the output current of the signal generator.
DataStudio calculates the back EMF, e = V − iR, and displays the results of the measurements
and calculations. In all the experiments, only two parameters are to be recorded, the current in
the rotor winding and the rotation velocity. Other results are calculated from these data.

4. Measurements and results

4.1. The initial periods

In this experiment, DataStudio records the initial periods of rotation for different voltages
applied to the motor. The signal generator operates in the auto mode: it creates the dc output
voltage only after starting a run. The sample rate is 10 Hz, and each run lasts 5 s; the option
automatic stop is used for this aim. The measurements are carried out for several values of
the applied dc voltage, from 1 to 4 V. Though the calculations of e are only valid under the
steady-state conditions, the records include data for transient conditions.
The data for voltages 2 and 4 V are presented here (figure 3). The graphs show how
the current in the rotor winding continually decreases until the rotor attains a steady rotation
velocity. For the unloaded motor, the steady current is nearly the same for both voltages
because the necessary torque should only balance the friction forces. For the loaded motor,
the steady rotation becomes slower, and the steady current increases. Second, the transient
Experiments with a dc motor 867

Figure 4. Records of initial periods of the motor operations for different loads. When the load
increases, the rotation velocity and the back EMF decrease.

period needed for establishing the steady state becomes shorter. The rotational inertia of
the system is constant, and the change in transients is caused by loading the motor. When
the current in the rotor winding becomes significant, an additional torque brings the rotation
velocity to its steady value in a shorter time.

4.2. The steady state for different loads

In the next experiment, the initial periods of rotation are recorded for a fixed applied voltage,
4 V, but different loads controlled by the magnetic braking. The latter is varied stepwise
(figure 4). The steady values of the rotation velocity and of the current in the rotor winding
are used for the calculations.

4.3. The torque and mechanical power

The steady current in the rotor winding, the back EMF and the calculated total mechanical
power produced by the motor are plotted versus the rotation velocity (figure 5). The torque
is proportional to the current. Therefore, the linear dependence of the current on the rotation
velocity confirms the linear dependence of the torque on the rotation velocity. The proportional
fit for the back EMF is used according to the model assuming a constant magnetic field
produced by the stator and a constant resistance of the rotor winding. For heavy loads (that
is, large currents), these assumptions are doubtful. First, the magnetic field produced by the
winding may become significant. Second, the heating of the winding increases its resistance,
while the back EMF is calculated assuming a constant resistance. Therefore, the deviations
from the proportional fit are quite expectable. If the motor is mechanically caused to rotate, it
becomes a dc generator providing an open circuit voltage strictly proportional to the rotation
velocity. Such a generator can be used as a speedometer.
The total mechanical power P, including the friction losses, follows equation (3). For
fitting purposes, the point of origin (ω = 0, P = 0) was added to the measurement data. With
868 Y Kraftmakher

current (mA)
back EMF (V)
mechanical power (mW)

rotation velocity (rad s−1)

Figure 5. Current in the rotor winding, back EMF and total mechanical power for different loads
versus the steady rotation velocity.

this addition, the fit equation is P (mW) = 7.6ω − 0.049ω2 (ω is given in rad s−1). From the
linear term of the fit, Ts = 7.6 × 10−3 N m. From the quadratic term, the rotation velocity
for the unloaded motor should be ωn = 155 rad s−1, which is close to the observations. Close
values of ωn are seen from the linear fits for the current in the rotor winding and for the back
EMF. The torque developed by the motor and thus the current in the motor winding linearly
depend on the rotation velocity: T (N m) = 7.6 × 10−3(1 − ω/155), i (A) = 0.272(1−ω/157).
Neglecting the difference between the two ωn values, the torque is proportional to the current:
T (N m) = 0.028i (A). The fit parameters are shown in figure 5. The results obtained are in
reasonable agreement with the accepted model of the motor.
Experiments with a dc motor 869

friction losses (mW)

useful power (mW)
motor efficiency

rotation velocity (rad s−1)

Figure 6. Power of friction losses, useful mechanical power and efficiency of the motor versus the
steady rotation velocity.

4.4. The friction losses and motor efficiency

To determine the motor efficiency, it is necessary to take into account the friction losses. Their
power depends only on the rotation velocity and can be determined by measuring the quantity
P1 = i1e1 (the current in the rotor winding times the back EMF for the unloaded motor). The
rotation velocity is changed by varying the applied voltage. The data (figure 6) are fitted with
a linear fit, P1 (mW) = 6.6 + 0.4ω (rad s−1). With this relation, DataStudio calculates P1
values for the same rotation velocities that are observed for the loaded motor. The efficiency
of the motor equals

η = (P − P1 )/ iV , (5)
870 Y Kraftmakher

where the P and iV data are related to the loaded motor; for each calculation, all the data
are taken for the same rotation velocity. The maximum efficiency appeared to be nearly 0.6,
which is a reasonable value for low-power motors.

5. Conclusion

The experiments with a dc motor are well related to university courses on electricity and
magnetism and can be recommended as lecture demonstrations or laboratory works in
undergraduate laboratories. One can see how the motor approaches its steady rotation and
how its mechanical properties depend on the rotation velocity. For characterization of the
motor, only two quantities are to be measured, the current in the rotor winding and the rotation
velocity. The results obtained confirm the accepted model of the motor.
Setups with a dc motor can be used for determining the viscosity of liquids. Like
the magnetic braking, the viscous braking is proportional to the rotation velocity. Various
modifications of the so-called rotational viscometer are widely used for practical needs.
Experiments with such viscometers were proposed for student laboratories (Hibberd 1952,
Courbin et al 2005). Experimentation with such devices seems to be a good theme for student
projects.

Acknowledgment

Many thanks to the referee for valuable comments and recommendations.

References
Bobrow L S 1996 Fundamentals of Electrical Engineering 2nd edn (New York: Oxford University Press)
Courbin L, Cristobal G, Winckert M and Panizza P 2005 Design of a low cost Zimm–Crothers viscometer: from
theory to experiment Am. J. Phys. 73 851–5
Gluskin E 1998 A possible way to measure the mechanical torque developed by an electric motor Eur. J. Phys. 19 93–5
Hecht E 1994 Physics (Pacific Grove, CA: Brooks/Cole)
Hewitt P 2003 Overheating motor Phys. Teach. 41 266–7
Hibberd F H 1952 A simple arrangement for a rotating cylinder viscometer Am. J. Phys. 20 134–5
Jewett J W and Serway R A 2008 Physics for Scientists and Engineers with Modern Physics 7th edn (London:
Brooks/Cole)
Kraftmakher Y 2001 Two experiments with rotating magnetic field Eur. J. Phys. 22 477–82
Lyna, Chia T C, Chia L S and Goh N K 2007 Demonstrating direct current through a dc motor with Pasco interface
datalogger Phys. Educ. 42 20–1
Ng P-h, Wong S-l and Mak S-y 2009 Efficiency measurement using a motor–dynamo module Phys. Educ. 44 639–43
Smith R J 1984 Circuits, Devices, and Systems 4th edn (New York: Wiley)
Turner L 2009 A simple demonstration of back emf Phys. Teach. 47 513–5
Whitaker R J 2009 Motors and bulbs in series Phys. Teach. 47 353–4
Yap J and MacIsaac D 2006 Analysing simple electric motors in the classroom Phys. Educ. 41 427–31