Tekhnê c Universidad Distrital Francisco José de Caldas

January - June 2018, Vol. 15, No. 1, pp. 13 – 20 ISSN 1692-8407

Mathematical modeling of a DC motor with separate excitation

Modelo matemático de un motor de corriente continua con excitación separada

Carlos A. Reyes G. Gustavo S. Gerena T.

Universidad Distrital Francisco José de Caldas Universidad Distrital Francisco José de Caldas
[email protected] [email protected]

This article proposes a model for a direct current motor with separate excitation using speed
control by armature current. Based on an understanding of its electromechanical functioning
and an adequate conceptualization of both the laws of physics and mathematical structure, this
model is constructed as a fundamental exercise for the development of control and simulation
schemes. To evaluate the performance of the model, performance curves are presented for a
test machine.

Keywords: Control, DC motor, dynamic systems, electromagnetism, electromechanical,

modeling, separate excitation

Este articulo propone un modelo para un motor de corriente continua con excitación
separada que utiliza control de velocidad por corriente de armadura. A partir de la
comprensión de su funcionamiento electromecánico y una adecuada conceptualización tanto
de leyes de la física como de la estructura matemática, se construye dicho modelo como
ejercicio fundamental para el desarrollo de esquemas de control y simulación. Para evaluación
del desempeño del modelo se presentan curvas de comportamiento para una máquina de
prueba.

Palabras clave: Control, electromagnetismo, electromecánico, excitación separada,

modelado, motor CC, sistemas dinámicos

Article typology: Research

Date manuscript received: December 6, 2017

Date manuscript acceptance: December 15, 2017

Research funded by: Universidad Distrital Francisco José de Caldas.

How to cite: Reyes, C., Gerena, G. (2018). Mathematical modeling of a DC motor with separate excitation.
Tekhnê, 15(1), 13 -20.

13
 Introduction motor, as it allows a greater range and degree of control over
speed and torque.
The DC motor is, in short, a torque transducer that In fact, the development of DC motors continues to
transforms electrical energy into mechanical energy (Kuo, advance, as evidenced, for example, by the work of electrical
2002). It is an important source of driving energy in today’s engineer and renowned inventor Frank Julian Sprague,
industrial and technological world, and it also outperforms who was recently honored in The History column in the
alternating current motors due to its high controllability of November-December 2015 issue of IEEE Power & Energy
speed and torque (Gordillo & Martínez, 2018). journal (Sprague, 2015). Sprague’s work pioneered the
The superior performance of the DC motor means that invention and development of constant speed DC motors,
it makes it possible to more easily perform some functions which, in addition to being non-slip, are completely
that the AC motor does not achieve (Martínez & Galvis, self-regulating, even under variable loads (Kuo, 2002). This
2006). Two examples of this are the DC motor’s ability to certainly expands the frontier of possibilities for the DC
develop torque several times greater in magnitude than an motor, and also justifies the objective of this paper.
AC motor of a similar size and the ability to operate at speeds One of the critical points of the DC motor has been
unreachable by an AC motor. the brushes collector system since the wear of the brushes
This document proposes an approximate mathematical due to friction, maintenance and the associated cost
model of the DC motor for this purpose of great utility in have reduced the efficiency of the machine. Currently,
both education and research (Martínez, Montiel, & Jacinto, however, not only are new prototypes of brushless
2016). We construct an equivalent circuit of the device, and DC motors (BLDC motors) being worked on, but
then we define the set of mathematical expressions that allow the effectiveness of automatic adjustment with PID
linking the physical variables that constitute the proposed controllers (Pongfai & Assawinchaichote, 2017) has
model. The constant values of the model were taken from even been increased by applying artificial intelligence
articles with models evaluated for the same type of electric (AI) algorithms (García, Osuna, & Martínez, 2018;
machine in order to compare the result obtained (Alvarez, Montiel, Martínez, & Jacinto, 2017).
2012).
Once the differential expression that relates the variables Description and principle of operation of the DC motor
of interest has been developed, they are taken to the
frequency domain through the application of the Laplace In the following, we present a schematic description of the
transform. Then the expression for the angular velocity ω engine and a synthesis of its basic operating principle.
is solved, the constants are replaced by the reference values, As can be seen in Fig. 1, the direct current motor is a
and finally, the inverse Laplace transform is applied to return machine composed of two parts, rotor A which is a sweet
to the time domain. steel cylinder mounted on a shaft that can rotate on its axis,
The resulting expression will then be the solution of the and stator or casing M where the permanent magnets or
proposed model, and will, therefore, describe the behavior of electromagnets are located if there are field coils F and F’. In
the angular velocity of the motor as a function of time. any case, the stator or inductor guarantees a magnetic field
In the second part of the article, graphical information between P and P’ as indicated by the dotted lines.
on the behavior of angular velocity as a function of time It should be pointed out that these field lines describe
is obtained using the SIMULINK simulator, which offers a the trajectories indicated as long as the rotor turns
graphical user interface (GUI) useful enough to build models counterclockwise as a result of the interaction between the
and to examine how successful the model developed has magnetic field of the inductor and the electric field of the
been. The final part of the article shares the results and builds armature. This dynamic happens because a current enters
conclusions. through the brushes or carbons to the brush collector and
circulates through the C conductors.
Background and current status of the DC motor Fig. 2 shows the front and side views of rotor A in Fig. 1.
Note that Fig. 1 does not show either the brush collector or
The Nikola Tesla Alternating Current (AC) induction the brushes for simplicity. The magnetic flux φ comes from
motor can be considered as the cornerstone of the triumph the stator.
of alternating current over the direct current system (Brittain, The first step in modeling a system, in this case, the DC
2005). However, there is one aspect where the contest does motor, is to understand its operating principle, so below we
not seem to have shifted to alternating current, and that is make a brief description of it.
what has allowed direct current motors to be so important By means of an excitation source va , an electric charge is
today. Observability and controllability are two essential circulated through the conductor of the armature, generating
factors in the analysis of dynamic systems, the latter being an ia current. This current gives rise to an electric field
the most important advantage of the DC motor over the AC that gives rise to a magnetic field, this phenomenon is

14
Tekhnê c Universidad Distrital Francisco José de Caldas
January - June 2018, Vol. 15, No. 1, pp. 13 – 20 ISSN 1692-8407

Once the principle of operation is understood the next step

is to elaborate the approximate equivalent circuit taking into
account that:
• The armature winding can be modeled as a series RL
circuit.
• In the rotor, a fem or force counter electromotive vb is
induced according to the Faraday induction law (Kuo, 2002).
• A direct voltage source va is applied to the armature
terminals.
The Kirchhoff tension law can then be applied to construct
the first expression that relates the variables va , Ra , La , and
vb , to the equivalent circuit shown in Fig. 3.

Figure 1. Schematic diagram of the operation of a DC motor

(Kuo, 2002). Figure 3. Equivalent circuit of the motor system in direct
current, separately excited.

Where:
• Ra represents the natural resistance of the conductor
forming the armature winding in ohms.
• La is the value of the armature winding inductance in
henrys.
• va is the DC voltage applied to the armature circuit in
volts.
• ia denotes the current flowing through the armature
conductor in amperes.
• vb represents the counter-electromotive force in volts.
• φ represents the stator’s fixed magnetic field.
• θ represents the angular displacement of the rotor in
Figure 2. Schematic diagram of a DC motor. radians.
• ω is the rotor angular velocity in rad/s.
• Ja denotes the equivalent rotational inertia of the rotor
shaft in kilograms.
described by Maxwell’s equations (Kuo, 2002). When the • β is the coefficient of viscous friction in N·m·s.
rotor field interacts with the magnetic field provided by the • τ f is the friction force or viscous damping in Newtons.
permanent magnet, the rotation of the armature is generated, • τe is the electromagnetic torque in Newtons.
a movement called torque. This phenomenon is described • τc is the resulting torque in Newtons.
by the Lorentz force equation which says that a conductor • k p is the electromagnetic torque constant in N·m/A.
carrying current experiences a force that tends to move it • kb is the counter electromotive force constant.
when placed in a magnetic field (Kuo, 2002) and by Lenz’s
Law which states that the direction of an induced current is Analysis for choice of variables and construction of the
such that it opposes the cause that produces it (Kuo, 2002). mathematical model
It is interesting that according to the above it is possible to
infer that the DC motor operates under electromechanical Making use of Kirchhoff’s voltage law in the equivalent
principles and that the torque that is a mechanical variable circuit shown in Fig. 3 relating to va , Ra , La and vb can be
can be controlled from electrical variables. obtained:

15
 d 2 ω(t) dω(t)
va (t) = vRa (t) + vLa (t) + vb (t) (1) dia (t) Ja · dt +β· dt
= (11)
dt kp
Expressing Eq. 1 in terms of the current:
Which is the second differential equation of the system,
dia (t) and describes the mechanical part of it. Finally, to obtain
va (t) = Ra ia (t) + La + vb (t) (2)
dt the differential equation that integrates the electrical and
The value of the counter electromotive force vb is defined mechanical parts, Eqs. 10 and 11 are substituted in Eq. 4,
as shown in Eq. 3. Replacing this expression in Eq. 2 results from where it results:
in Eq. 4:
 dω(t)   d2 ω(t) dω(t)

 Ja dt + β · ω (t)   J +β·
+La  a dt

vb (t) = kb ω (t) (3) dt
va (t) = Ra  +kb ω (t)
kp  kp 
(12)
dia (t)
va (t) = Ra ia (t) + La + kb ω (t) (4) It is important to note that this expression relates
dt to electrical variables such as current and voltage with
Then Eq. 4 is the first differential equation of the model mechanical variables such as torque and friction force, which
and describes the electrical part of the system. is consistent because it is an electromechanical system in
For the description of the mechanical part, it is necessary itself.
to find the second differential expression. As it is a linear Eq. 12 describes the mathematical model of the separately
system, it is assumed that the torque developed by the motor excited DC motor. Before proceeding with the solution of
is proportional to the flow between the iron and the armature the mathematical model it must be considered that Eq. 12 is
current. In other words, it is possible to conceptualize an a second order differential equation, but it must also be taken
equation of torques that relates the applied force or resultant into account that the value of L approaches zero in direct
torque τc , the inertia of the armature τe and the friction force current motors with independent excitation, so Eq. 12 can
τ f which, said to be a step, depends on the angular velocity be simplified to a first-order homogeneous, linear differential
ω, and opposes the movement. So: equation of constant coefficients.
 dω(t) 
τc (t) = τe (t) − τ f (t) (5)  Ja dt + β · ω (t) 
va (t) = Ra    + kb · ω (t) (13)
kp
In addition τc is described by the expression:

dω (t) Mathematical model solution

τc (t) = Ja · (6)
dt In order to be able to apply the transform of Laplace, it
The resulting torque τc is generated by the is necessary to establish that ω(0) = 0. Then Eq. 13 is
electromagnetic torque τe , which in turn depends on transformed as follows:
the armature current ia and is defined as:
" # " #
Ja R a dω (t) β · ω (t)
τe (t) = k p · ia (t) (7) L [va (t)] = ·L +Ra ·L +L [Kb · ω (t)]
kp dt kp
Where k p is the electromagnetic torque constant. The (14)
viscous damping τ f , or viscous friction force, is defined as:
Va (s) Ja · Ra Ra · β
τ f (t) = β · ω (t) (8) = · s · W (s) + W (s) + kb · W (s) (15)
s kp kp
Now, replacing Eqs. 6, 7 and 8 in Eq. 5 gives: Factoring W(s):
dω (t) Va (s)
"
Ja · R a Ra · β
#
Ja · = k p · ia (t) − β · ω (t) (9) = W (s) · ·s+ + kb (16)
dt s kp kp
Solving for ia (t) in Eq. 9, results:
By doing:
dω(t)
Ja · dt + β · ω (t) Ja · R a Ra · β
ia (t) = (10) = λ and + kb = δ (17)
kp kp kp
Then, deriving the Eq. 10 is obtained: And rewriting Eq. 16 comes out:

16
Tekhnê c Universidad Distrital Francisco José de Caldas
January - June 2018, Vol. 15, No. 1, pp. 13 – 20 ISSN 1692-8407

Table 1
Va (s) Experimental data of the DC motor.
= W (s) · (λs + δ) (18)
s
Solving for W(s):

Va (s)
W (s) = (19)
s (λs + δ)
The partial fraction method is now applied to Eq. 19.
" #
A B
W (s) = Va (s) · + (20)
s (λs + δ)

" # " #
1 A B
Va (s) · = Va (s) · + (21)
s (λs + δ) s (λs + δ) Finally, we return the replacement of δ and λ and replace
! ! the values of Ja , Ra , kb , β, and k p supplied by the reference
1 A B article (table 1).
= + (22)
s (λs + δ) s (λs + δ)  Ra ·β 
+kb
kp

va (t)  Ja ·Ra t 

1 = A (λs + δ) + B · s (23) ω (t) = R ·β 1 − e kp
 (32)
a
k p + kb
 

1 = Aλs + Aδ + B · s (24) Finally Eq. 33 is obtained, after replacing the parametric

values given in the reference article.

s · (0) + 1 = s (Aλ + B) + Aδ (25)

h i
ω (t) = 111.99 · 1 − e−1.4287t (33)
Then we get that:
Simulations
1
1 = Aδ =⇒ A = (26) Below are graphs of torque, angular velocity, armature
δ
current, and counter-electromotive force obtained with the
λ Simulink simulator. The construction was done using block
0 = Aλ + B =⇒ B = −Aλ =⇒ B = − (27) diagrams (Figs. 4, 5 and 6, Fig. 4 shows the block diagram
δ
used in the simulation, Fig. 5 shows the block diagram for
And Eq. 20 becomes:
angular velocity, and Fig. 6 shows the block diagram for
1
 δ − λδ 
 armature current) from the differential relationships obtained
W (s) = Va (s) ·  +  (28) by solving di(t)/dt in Eq. 4 and dω(t)/dt in Eq. 9.
s (λs + δ)  Fig. 7 shows the behaviour of the angular velocity as a
Dividing between λ numerator and denominator of the function of time.
second term of the bracket, and factoring 1/δ to bring the Fig. 8 shows the behavior of the armature current as a
expression to a known form we obtain: function of time.
  Fig. 9 shows the torque as a function of time.
Va (s)  1 1  Fig. 10 shows the counter-electromotive force as a
W (s) = ·  −    (29) function of time.
δ s s + λδ
Now Laplace inverse transform is applied to return to the Conclusions
time domain and obtain the Eq. 31 which is the solution of
The mathematical model of the DC motor with
the model in the time domain.
independent excitation can be described by the expression:
 Ra ·β 
  +kb
" # kp
va (t) va (t) t
−1 1
 1  
va (t)  Ja ·Ra

−1 −1 
L [W (s)] = ·L − · L     (30)

ω (t) = R ·β 1 − e kp
(34)
δ s δ

s+ δ λ
a
k p + kb
 

va (t) h δ i
This equation can be easily implemented in simulation
ω (t) = 1 − e− λ t (31) software such as Simulink for specific values of machine
δ

17
Figure 4. Diagram of blocks and subsystems used in the simulation.

Figure 5. Angular velocity subsystem diagram.

Figure 6. Block diagram for armature current.

parameters, and derive with respect to time the behavior of for the same machine parameters, validating the behavior of
both the angular velocity and the armature current and torque the model.
developed. It is even possible to consider more complex
situations, for example, to vary the value of the armature
resistance over time as if it were a variable resistance in References
the machine, or to vary the armature voltage to analyze
the behavior of a speed control scheme. The test curves Alvarez, M. (2012). Modelo matemático de un motor de
developed for the machine matched those reported in articles corriente continua separadamente excitado: Control de
velocidad por corriente de armadura. Latin-American
Journal of Physics Education, 6(1), 155-161.

18
Tekhnê c Universidad Distrital Francisco José de Caldas
January - June 2018, Vol. 15, No. 1, pp. 13 – 20 ISSN 1692-8407

Figure 7. Angular speed.

Figure 8. Armature current.

Figure 9. Torque as a function of time.

Brittain, J. (2005). Electrical engineering hall of fame: Engineering Sciences, 11(39), 1945-1952.
Nikola tesla. Proceedings of the IEEE, 93(5), Kuo, B. (2002). Automatic control systems (Vol. 8). Wiley.
1057-1059. Martínez, F., & Galvis, J. (2006). Control escalar en motores
García, C., Osuna, B., & Martínez, F. (2018). Safe grasping de inducción monofásicos. Tecnura, 10(19), 29-37.
system for robotic hand with fingers. Contemporary Martínez, F., Montiel, H., & Jacinto, E. (2016).
Engineering Sciences, 11(40), 1991-1998. Inductive teaching and problem-based learning as
Gordillo, J., & Martínez, F. (2018). Fuzzy control for bipedal significant training tools in electrical engineering.
robot considering energy balance. Contemporary Smart Innovation, Systems and Technologies, 59(1),

19
Figure 10. Counter-electromotive force as a function of time.

179-188. control systems based on artificial intelligence. In

Montiel, H., Martínez, F., & Jacinto, E. (2017). International electrical engineering congress (ieecon
Visual patterns recognition in robotic platforms 2017) (p. 1-4).
through the use of neural networks and image Sprague, J. (2015). A sprague invention: Multiple unit train
processing. International Journal of Applied control [history]. IEEE Power and Energy Magazine,
Engineering Research, 12(18), 7770-774. 13(6), 88-103.
Pongfai, J., & Assawinchaichote, W. (2017). Optimal
pid parametric auto-adjustment for bldc motor

20