LATEST TRENDS on ENERGY & DEVELOPMENT, ENVIRONMENT & BIOMEDICINE

Brushless D.C. Motor For Medical Applications

ABDEL-KARIM DAUD
Electrical and Computer Engineering Department
Palestine Polytechnic University (PPU)
P.O. Box 198 Hebron, Tel:+972-2-2228912, Fax:+972-2-2217248
PALESTINE
Email: daud@ppu.edu
Abstract:- The paper deals with a controlled brushless d.c. motor (BLDCM) for a left ventricular heart assist
system. The medical system consists of a bloodpump, a centrifugal pump, a brushless d.c. motor, a solenoid,
power supply and controls. The brushless d.c. motor operates sensorless and is developed with respect to the
special requirements of an implantable heart assist device. Here should be investigated the effect of rotor
magnet design on artificial heart driving motor performance, in which rotors with different magnet lengths or
thicknesses, as well as different peripheral angles, were designed and simulated in the same motor stator with
different rotating speeds. This simulation is based on the calculation of the air-gap flux density of the motor,
which is obtained by using three dimensional analytical field method. Then the motor efficiency was
computed. MATLAB was used to implement the procedures to calculate the required parameters. The results
demonstrated that the reduction of rotor magnet size and the enlargement of the air gap between the rotor
magnets and the stator coil core have no significant effect on motor efficiency, but will reduce the torque value
on which the motor achieves the highest efficiency; it could be remedied however by increasing the rotating
speed, because the torque at the high efficiency point will increase along with the rotating speed. These results
may provide a basis for developing small rotor magnets, large air gap and high efficiency motors for driving
an artificial heart pump.
Key-Words:- Heart Assist System, BLDCM, Permanent Magnets, Sensorless Control, MATLAB
successfully in designing a new mini motor for an
artificial heart pump device.
Permanent-magnet excited brushless dc motors
(BLDCM) are becoming increasingly attractive in
many applications due to their performance
advantages such as reduced size and weight, high
efficiency, low noise and maintenance, improved
reliability and very good control characteristics in a
wide speed range [8,10,13,14].
The most popular brushless DC motors have threephase windings [6] or four-phase windings [7],
which are controlled and driven by full bridge
transistor circuit [9].
From these classical types of brushless dc motor
with three or four windings, it became clear that the
electronic circuit is still expensive so that this type
of motor could not be used in consumer
applications.
In this paper, a brushless DC motor with two air
gap windings and PM-rotor in different
arrangement is investigated [5]. Therefore, the cost
of the electronic circuit is reduced.
Elimination of position and velocity transducers in
dc drives is desirable for a number of reasons. The
shaft transducers themselves and perhaps even

1 Introduction
The artificial heart (or heart assist system) finds
more and more applications and has been the
subject of much research in recent years. Usually, it
is a type of centrifugal blood pump driven by a
motor. There are left ventricular and right
ventricular heart assist systems available. A left
ventricular assist device pumps the blood from the
left apex into the aorta and could be helpful in 80%
of cases, where heart transplantation is necessary.
Because of risk of infection a heart assist device
have to be totaling implantable. Therefore a highly
durable and reliable system with smallest possible
volume and weight is required. The left ventricular
heart assist system is made up of a blood pump, a
centrifugal pump, a brushless d.c. motor, a
solenoid, power supply and controls [1, 2,3]. The
highly efficient brushless d.c. motor drives the
centrifugal pump.
In order to reduce the size and weight of the pump,
improvement of the motor design was of greatest
importance. The simplest way of minimizing the
motor is to reduce the magnet size of the rotor and
to investigate its effect on the torque and the
efficiency of the motor. The results are applied

ISSN: 1792-4340

27

ISBN: 978-960-474-209-7

LATEST TRENDS on ENERGY & DEVELOPMENT, ENVIRONMENT & BIOMEDICINE

more the associated wiring, are a significant source

of failure and cost. However, for a good
performance of the PM dc motor, the rotor position
has to be known with high accuracy. Much effort,
reflected in the numerous papers dealing with this
issue, has been made in research to avoid use of a
position sensor in dc drives [11, 12]. In this paper,
the detection of back-emf technique is used.
The air gap flux density is obtained by using a three
dimensional analytical method [4]. By using
MATLAB, the different motor parameters and
characteristics are determined.

2 Motor Design and its Drive Circuit

Fig.2: Drive circuit of two phase brushless DC

motor

A total implantable system requires a drive with

small volume and weight. The specific parameters
are: 824 kW/m3 and 5896 Nm/m3. The bloodpump
operates at high speeds (5000 ... 7000 rpm)
combined with high torque (15 ... 20 mNm) [1]. The
developed motor with a very small mass of magnets
and its air-gap windings operates in the required
high speed- and torque-range. For an optimal
design of the motor an analytical method has been
developed. The aim is to find out the best design
parameters for the optimal use of the available
volume and to reach the highest efficiency at the
systolic operating point (6000 rpm, 15 mNm).
The block diagram of a two-pole, two-phase
brushless dc motor drive is shown in Fig.2, while
the basic construction of this motor is shown in
Fig.1.

The electromagnetic structure of the drive system,

as shown in Fig.1, is formed of an internal rotor
ring formed permanent magnet arrangement, and a
slotless stator assembly that consists of a lamination
stack with air gap windings. The construction and
distribution of stator windings are shown in Figs.3
and 4. The motor has two phases. Each phase
consists of four coils. These coils combine two
parallel groups. Each group consists of two series
coils at the same radius as shown in Fig.3.
Therefore, the first phase has the two parallel
groups: (A0.0, B0.0) and (A1.0, B1.0), while the
remaining coils (A0.5, B0.5) and (A1.5, B1.5)
combine the second one.

stator
winding
rotor

airgap

PM

Fig. 1: Construction of BDCM

Fig. 3: Construction of stator windings

ISSN: 1792-4340

28

ISBN: 978-960-474-209-7

LATEST TRENDS on ENERGY & DEVELOPMENT, ENVIRONMENT & BIOMEDICINE

R3

R2

R1

Fig. 6: Rotor magnets with radial thickness

90

Fig. 4: One group-phase connection

120

180

Fig.7: Rotor magnets with different peripheral

angles

In this case, the current will flow in each phase for

a period of about (180) for every rotor revolution.
An increase in the number of phases will only
provide a slight increase in the average torque, but
it requires more power switches for the electronic
commutation circuit. The power converter, as
shown in Fig. 2, is a double bridge circuit
(inverter) that provides the required sequence of
the phase voltages in both directions [9]. The
powers MOSFETs receive their gate signals from
the microprocessor through the 8155 PID interface
circuit. Detection of the rotor position can be
achieved through photocell or sensorless control
circuit.

Several rotor arrangements were considered. They have

different axial lengths of 24 mm (L24), 16 mm (L16)
and 8 mm (L8) respectively (Fig. 5), or different radial
thickness of 3 mm (R3), 2 mm (R2) and 1 mm (R1)
separately (Fig. 6), as well as different peripheral
angels of 180 (180),120 (120) and 90 (90)
(Fig.7). Therefore, a rotor described as L24R3180
means that this rotor magnet has 24 mm axial length, 3
mm radial thickness and 180 peripheral angles. Many
rotor magnets were used in the simulation; they are
L24R3180, L24R3120, L24R390, L16R3180,
L16R3120, L16R390, L24R2180, L24R1180
and L8R3180.
According to Fig. 6, the air gap between the rotor
magnets and the stator coil core can be enlarged 1mm
radially by use of R2 and 2 mm by use of R1 and
reducing the rotor radius, as compared with R3. That is
to say, it is necessary to consider the effect of both the
rotor magnet size and the air gap on motor performance
during changes of the rotor magnet thickness.
These rotors were simulated in the same motor stator
coil.

L24

L16

3 Air Gap Flux Density

The air-gap flux density B is obtained from a
three dimensional analytical method in the author'
field calculation, which is based on Bio-Savaratlaw by using mirror laws under the influence of the
iron yoke [4].

L8

4 System Model

Fig.5: Rotor magnets with different axial length

ISSN: 1792-4340

29

ISBN: 978-960-474-209-7

LATEST TRENDS on ENERGY & DEVELOPMENT, ENVIRONMENT & BIOMEDICINE

Fig. 8 shows the equivalent circuit for one phase.

The angle in Fig.8 will give the actual position
of the rotor. During 1 2 the winding is
energized from the dc source with the voltage V, in
which the induced voltage is positive as shown in
Fig.9.
Therefore, the transistor switches on this period,
which is called a current flow angle and given
by
= 2 1

Fig.9: Phase current as function of rotor position

for L
0

(1)

where 1 and 2 are switch-on and -off angles

respectively. For the electronic switch a voltage
drop of VT is assumed. The curve of transistor
voltage VT is represented in Fig.10.

winding
2

1
2

Fig.10: Transistor voltage as function of rotor

position for L
0

By inspection of Fig.8:
V=e+Ri+L

rotor

rotor
rotor

di
+ VT for 1 2
dt

(2)

With

stator

d
d
=
,
(3)
dt
d
the governing equation can be arranged into the
following form:
di
Ri()+L
=V-e()-VT() for 1 2(4)
d

=t
airgap
winding
(a)

and

5 Voltages and Currents

The magnetic flux density is obtained by
applying the 3 dimensional field methods. Each
winding is characterized by number of windings,
average diameter of winding, active length and
winding parameter.
Therefore the induced voltages can be written as
e() = Cm B()

where Cm is the motor constant.

Taking into account eqs. (4) and (5) and for steady
state conditions, the instantaneous current can be
obtained. Then the average value of the armature
current can be calculated over the period (0, 2)
and is given by:

(b)
Fig.8: Drive circuit configuration of the motor

ISSN: 1792-4340

(5)

30

ISBN: 978-960-474-209-7

LATEST TRENDS on ENERGY & DEVELOPMENT, ENVIRONMENT & BIOMEDICINE

Iave =

1
2

0.6

2 N

i j( )d

(6)

j=1

0.5

where N is the number of armature phases.

0.4

B/T

6 Torque

0.3

The gross electromagnetic torque is given by [10]:

T = Cm B() i()

0.2

(7)

0.1

In order to produce a continuous torque, two or

more phases have to be energized in sequence.
Therefore, the average torque Tave can be
calculated over the period (0, 2 ) during the
current flow angle o.
The efficiency of the drive system is
=

Tave
V Iave

50

100

150

/
Fig. 11: B distribution for different magnet length
0.6

(8)
0.5

7 Simulation and Results

0.4

The previous described method has been applied to

a two pole, two phase brushless DC motor with air
gap
windings
and
different
PM-rotor
arrangements. The rotor consists of a high energy
sintered Nd-Fe-B magnet material. The special
feature of this brushless DC motor is that no rotor
position sensors, such as Hall probes or
magnetoresistive sensors, are used. For detection
of rotor position the induced voltage is used in
combination with an electronic circuit. A three
dimensional analytical method was used for field
calculations.
The simulation of the transient and steady-state
characteristics of the motor was carried out using
the well-known software package MATLAB. The
instantaneous air-gap flux density B is shown in
Figs.11, 12 and 13. The rotor magnet length has
significance only for low values of magnet length
(L=8mm) according to Fig. 11. While different
magnet peripheral angles have effect on the shape
and values of air-gap flux density B as shown in
Fig. 12. It is clear; the values of B are smaller by
decreasing the angle. The rotor magnet radial
thickness has a great effect on the flux density
distribution B as shown in Fig. 13.The shape and
values of B are lower by decreasing magnet
thickness for constant air gap length. That means,
these values will be very lower by enlarging the air
gap length.

ISSN: 1792-4340

L24R3180
L16R3180
L8R3180

B/T

L24R3180
L24R3120
L24R390

0.3
0.2
0.1
0

50

100
/

150

Fig. 12: B distribution for different magnet

peripheral angles
0.6
0.5
0.4

B/T

0.3
0.2

L24R3180
L24R2180
L24R1180

0.1
0

100
150
/
Fig. 13: B distribution for different magnet radial
thickness

31

50

ISBN: 978-960-474-209-7

LATEST TRENDS on ENERGY & DEVELOPMENT, ENVIRONMENT & BIOMEDICINE

Fig. 14 demonstrates the effect of the rotor magnet

axial length on the motor torque and the efficiency.
The rotor magnet length has no significance on the
efficiency of the motor according to Fig. 14, only
the torque achieving the highest efficiency has
decreased together with the rotor magnet length.
Equally, different magnet peripheral angles have
also no obvious effect on motor efficiency (Fig.
15), only the torque achieving the highest
efficiency has decreased together with the rotor
magnet peripheral angle.

The rotor magnet radial thickness again has an

unremarkable effect on the motor efficiency
according to the curves of L24R3180 and
L24R2180 (Fig. 16). The efficiency of the
L24R1180 curve has decreased, indicating that
the rotor magnets could not be too thin and the air
gap could not be too big.
It seems that the rotating speed has a very small
effect on motor efficiency (Fig. 17), only the
torque of the highest efficiency point increases
together with the rotating speed.

0.7

0.7

0.6

0.6
L8R3180
L16R3180
L24R3180

0.5

0.4

0.5
0.4

n = 6500rpm

0.3

0.3

0.2

0.2

0.1

0.1

L24R3180
L24R2180
L24R1180

n = 6500rpm

0.01

0.03

0.02
T / Nm

0.01

0.02

0.03

0.04

T / Nm

Fig. 14: Effect of rotor magnet axial length on

motor torque and efficiency

Fig. 16: Effect of rotor magnet radial thickness on

motor torque and efficiency

0.7

0.7

0.6
0.6
0.5

0.5

0.4

L24R3180
L24R3120
L24R390

0.3

0.4
0.3

0.2
n = 6500rpm

0.01

0.02

0.03

0.04

L24R3180

0.2

0.1
0

n=5000rpm
n=5500rpm
n=6000rpm
n=5000rpm

0.1

0.05

T / Nm

0.01

0.02

0.03

0.04

T / Nm

Fig. 15: Effect of rotor magnet peripheral angles

on motor torque and efficiency

ISSN: 1792-4340

Fig. 17: Effect of rotating speed on motor torque

and efficiency

32

ISBN: 978-960-474-209-7

LATEST TRENDS on ENERGY & DEVELOPMENT, ENVIRONMENT & BIOMEDICINE

WSEAS Transactions on Power Systems, May

2006, Issue 5, Volume 1, pp. 802-809.
[6] A.-K. Daud; S.H. Khader, An analytical
method for determination of commutation leading
angle in three- phase brushless DC motor, Proc. Of
ELECTROMOTION97, Cluj-Napoca, pp. 253258.
[7] R. Hanitsch, Analog computer simulation of
brushless DC motors, Proc. of ICEM, 1984,
Lausanne, pp. 189-192.
[8] J.M.D. Coey, Rare-Earth Iron Permanent
Magnets, Oxford Science Publications, Clarendon
Press Oxford, 1996.

8 Conclusions
This paper presented a performance analysis for a
brushless DC motor with two phases for medical
application. This motor type has air gap windings
and different arrangements of Nd-Fe-B PM-rotor.
For calculation the air-gap magnetic flux density, a
three dimensional analytical method was used.
With results of air gap flux density, motor
characteristics were calculated. This analytical
method was used to reveal the effect of rotor
magnet size (axial length, radial thickness and
peripheral angle) and air gap on motor torque and
efficiency. It is certified that the rotor size and air
gap have no significant effect on motor efficiency,
but the motor torque value at which the highest
efficiency is achieved will increase along with
increasing motor size and along with decreasing
air gap. This means a small rotor and large air gap
motor could achieve a high efficiency, but its
output torque will reduce if the motor performs at
its highest efficiency point or close by.
Fortunately, this shortcoming could be remedied
by increasing the rotating speed of the motor,
because the high efficiency point has larger output
torque at higher speed than at lower speed. It is
concluded that a small rotor motor with large air
gap and large output torque and with high
efficiency is feasible, as long as the rotating speed
is high enough. The rotating speed is a very
important factor for blood pump design. The motor
design should therefore be performed with the
pump development simultaneously.

[9] M.H. Rashid, Power Electronics, Circuits,

Devices and Applications, Pearson Prentice
Hall, Upper Saddle River, New Jersy, 2004.
[10] H.B. Ertan; M.Y. ctung; R. Colyer; A.
Consoli, Modern Electrical Drives, Kluwer
Academic Publishers, Netherlands, 2000.
[11] L. Harnefors; P. Taube; H.-P. Nee, An
improved method for sensorless adaptive control
of permanent-magnet synchronous motors, Proc.
Eur. Conf. Power Electronics, Trondheim, Vol 4,
1997.
[12] N. Matsui, Sensorless permanent-magnet
brushless DC and synchronous motor drives,
ELECROMOTION, Vol 3, No. 4, 1996, pp. 172180.
[13] J.D. Ede; Z.Q. Zhu; D. Howe, Design
considerations for high-speed permanent magnet
brushless dc motors, Proc. International
Conference on Power Electronics, Machines, and
Drives, IEE, 2004, pp. 686-690.
[14] B.-K. Lee; M. Ehsani, Advanced simulation
model for brushless dc motor drives, Electric
Power Components and Systems, September 2003,
pp 841-868

References:
[1]
R. Hanitsch; N. Parspour, Contribution to
brushless dc motor design for a heart assist device,
ICEM, 1992, Manchester, Vol. 1 pp. 165-169.
[2]
A.C. Guyton; J. E. Hall, Textbook of
Medical Physiology, Philadelphia, Pa, Elsevier
Saunders, 2006.
[3] K. X. Qian; H. Y. Yuan; W. M. Ru; P. Zeng,
Technical note: Experimental method to reveal the
effect of rotor magnet size and air gap on artificial
heart driving motor torque and efficiency, Journal
of Medical Engineering & Technology, Sept./Okt.
2002, Number 5, Volume 26, pp. 199-201.
[4] A.-K. Daud, A three dimensional field analysis
of permanent magnets, The Palestinian
Engineering Conference, August 2001, Nablus, pp.
144-157.
[5] A.-K. Daud, Performance analysis of two phase
brushless DC motor for sensorless operation,

ISSN: 1792-4340

33

ISBN: 978-960-474-209-7