3 Permanent-magnet materials and

circuits

3.1 Permanent-magnet materials and characteristics

The sustained success of the permanent-magnet industry in developing

improved magnet characteristics is evident from Fig. 3.1, the latest addition
being neodymium-iron-boron which has been pioneered by Sumitomo as
'Neomax', General Motors as 'Magnequench', Crucible ('Crumax'), and IG
Technologies ('NelGT'). At room temperature NdFeB has the highest energy
product of all commercially available magnets. The high remanence and
coercivity permit marked reductions in motor framesize for the same output
compared with motors using ferrite (ceramic) magnets. However, ceramic
magnets are considerably cheaper.
Both ceramic and NdFeB magnets are sensitive to temperature and special
care must be taken in design for working temperatures above 100°C. For very
high temperature applications Alnico or rare-earth cobalt magnets must be
used, for example 2-17 cobalt-samarium which is useable up to 200°C or even
250°C.
NdFeB is produced either by a mill-and-sinter process (Neomax) or by a
melt-spin casting process similar to that used for amorphous alloys
(Magnequench). Powder from crushed ribbon is bonded or sintered to form

50-

40-

NdFcB
30-
o
Rare-Earth/Cobalt

20-

Alnico 9
10- Alnico 5-7 r
Alnico 5

1950 1960 1970 1980

Year

FIG. 3.1. Development of PM materials in terms of maximum energy product.

the MQI or MQII grades produced by Magnequench Division of GM. The
MQI bonded magnets can be formed in a wide variety of shapes. They are not
100 percent dense and coatings may be used to prevent corrosion. With MQII
and other sintered materials a dichromate coating may be used, or
electroplating.
For lowest cost, ferrite or ceramic magnets are the universal choice. This
class of magnet materials has been steadily improved and is now available with
remanence of 0.38 T and almost straight demagnetization characteristic
throughout the second quadrant. The temperature characteristics of ferrite
magnets can be tailored to the application requirements so that maximum
performance is obtained at the normal operating temperature, which may be
as high as 100°C.
A brief summary of magnet properties is given in Table 3.1. More detail can
be obtained from suppliers' data sheets, as the examples show. Specialist data
and measurements are often made by permanent-magnet research and
development bodies; for example, in the UK, the Magnet Centre at
Sunderland Polytechnic, and in the USA, the University of Dayton, Ohio.
Activity in magnet research is also well reported in IEEE and specialist
conference proceedings.

Table 3.1. Magnet properties

Alnico
Property Units 5-7 Ceramic Sm2Co17 NdFeB

Br T 1.35 0.405 1.06 1.12

T 0.074 0.37 0.94 1.06
(BH) max MGOe 7.5 3.84 26.0 30.0
/<rec . 1.9 1.1 1.03 1.1
Specific gravity 7.31 4.8 8.2 7.4
Resistivity /(O cm 47 >104 86 150
Thermal expan-
sion 10" 6 /°C 11.3 13 9 3.4
B, temperature %/°C -0.02 -0.2 -0.025 -0.1
coefficient
Saturation H kOe 3.5 14.0 >40 >30

3.2 B-H loop and demagnetization characteristics

The starting-point for understanding magnet characteristics is the B-H loop

or 'hysteresis loop', Fig. 3.2. The x-axis measures the magnetizing force Q£
'field intensity' H in the material. The y-axis is the magnetic flux-density Sjn.
the material. An unmagnetized sample has Z?=0and H = 0 and therefore starts
out at the origin. If it is subjected to a magnetic field, as for example in a
magnetizing fixture (an electromagnet with specially shaped pole pieces to
focus flux into the magnet), then B and H in the magnet will follow the curve
OA as the external ampere-turns are increased. If the external ampere-turns
are switched off, the magnet relaxes along AB. Its operating point (H, B) will
depend on the shape of the magnet and the permeance of the surrounding
'magnetic circuit'. If the magnet is surrounded by a highly permeable magnetic
circuit, that is, if it is 'keepered', then its poles are effectively shorted together
so that H=0 and the flux-density is then the value at point B, the remanence
Br. The remanence is the maximum flux-density that can be retained by the
magnet at a specified temperature after being magnetized to saturation.
External ampere-turns applied in the opposite direction cause the magnet's
operating point to follow the curve from B through the second quadrant to C,
and again if they are switched off at C the magnet relaxes along CD. It is now
 magnetized in the opposite direction and the maximum flux-density it can
retain when 'keepered' is — BT. To bring the flux-density to zero from the
original positive remanence the external ampere-turns must provide within
the magnet a negative magnetizing force — H Q, called the coercivity. Likewise,
to return the flux-density to zero from the negative remanence point D, the
field + H C must be applied. The entire loop is usually symmetrical and can be
measured using special instruments such as the Hysteresisgraph made by
Walker Scientific Instruments.
If negative external ampere-turns are applied, starting from point B, and
switched off at R, the operating point of the magnet 'recoils' along RS. If the
magnet is still 'keepered' the operating point ends up at point S. Now if the
external ampere-turns are re-applied in the negative direction between S and
R, the operating point returns along SR. The line RS is actually a very thin
'minor hysteresis loop' but for practical purposes it can be taken as a straight
line whose slope is equal to the recoil permeability. This is usually quoted as a
relative permeability, so that the actual slope of RS is n rec fx 0 H/m. Operation
along RS is stable provided that the operating point does not go beyond the
boundary of the original hysteresis loop.
A 'hard' PM material is one in which the hysteresis loop is straight
throughout the second quadrant, where the magnet normally operates in
service. In this case the recoil line is coincident with the second-quadrant
section of the hysteresis loop. This is characteristic of ceramic, rare-
earth/cobalt, and NdFeB magnets, and the recoil permeability is usually
between 1.0 and 1.1. 'Soft' PM materials are those with a 'knee' in the second
quadrant, such as Alnico. While Alnico magnets have very high remanence
and excellent mechanical and thermal properties, they are limited in the
demagnetizing field they can withstand. It should be noted that compared
with lamination steels even the 'soft' PM materials are very 'hard': in other
words, the hysteresis loop of a typical nonoriented electrical steel is very
narrow compared with that of even the Alnico magnets.
The most important part of the B-H loop is the second quadrant, drawn in
more detail in Fig. 3.3. This is called the demagnetization curve. In the absence
of externally applied ampere-turns the magnet's operating point is at the
intersection of the demagnetization curve and the 'load line', whose slope is the
product of ¡x0 and the permeance coefficient of the external circuit (see
Chapter 4, Section 4.2).
Since B and H in the magnet both vary according to the external circuit
permeance, it is natural to ask what it is about the magnet that is 'permanent'.
The relationship between B and H in the magnet can be written
B = ii0H+J.
The first term is the flux-density that would exist if the magnet were removed
and the magnetizing force remained at the value H. Therefore the second term
can be regarded as the contribution of the magnet to the flux-density within its
 Permeance coefficient
1.0 2.0

0.8

-0.6

ce

-0.2

-K.
0
2.0 1.5 1.0 0.5
- H . H (T)

FIG. 3.3. Second-quadrant demagnetization characteristic showing intrinsic curve.

own volume. Clearly if the demagnetization curve is straight, and if its relative
slope and therefore the recoil permeability are both unity, then J is constant.
This is shown in Fig. 3.3 for values of negative H up to the coercivity. J is called
the magnetization of the magnet. Obviously it has the units of flux density, T.
In most hard magnets the recoil permeability is slightly greater than 1 and
there is a slight decrease of J as the negative magnetizing force increases, but
this is reversible down to the 'knee' of the B-H loop (which may be in either the
second or the third quadrant, depending on the material and its grade.
The magnetization and other parameters of the linear or 'recoil line' model
are used in finite-element analysis for calculation of magnetic circuits, and
several commercial packages are available to do this.
Evidently the magnet can recover or recoil back to its original flux-density
as long as the magnetization is constant. The coercive force required to
permanently demagnetize the magnet is called the intrinsic coercivity and this
is shown as H ci.
Another parameter often calculated is the magnet energy product, which is
simply the product of B and H in the magnet. This is not the actual stored
magnet energy, which depends on the history or trajectory by which the
magnet arrived at its operating point and usually cannot be calculated except
under very artificial conditions. The energy product is a measure of the stored
energy but, more importantly, it gauges how hard the magnet is working to
provide flux against the demagnetizing influence of the external circuit.
Contours of constant energy product are rectangular hyperbolas and these are
frequently drawn on graphical property data sheets provided by magnet
Scanned by CamScanner
Scanned by CamScanner
Scanned by CamScanner
Scanned by CamScanner
Scanned by CamScanner
Scanned by CamScanner
Scanned by CamScanner
Scanned by CamScanner
Scanned by CamScanner
 BLDC square wave motor

4.3 Squarewave brushless motor: torque and e.m.f. equations

The basic torque and e.m.f. equations of the brushless d.c. motor are quite
simple, and resemble those of the d.c. commutator motor. The following
derivation attempts to encompass several fundamental aspects of these two
equations, so as to lay a foundation for understanding the control
characteristics and limitations, and the similarities and differences with other
machines.
A simple 'concept machine' is shown in Fig. 4.5(a). Note that the two-pole
magnet has a pole arc of 180 degrees, instead of the 120 degrees analysed in the
previous section. The airgap flux-density waveform is ideally a square wave as
shown in Fig. 4.5(b). In practice, fringing causes the corners to be somewhat
rounded. The coordinate axes have been chosen so that the centre of a north
pole of the magnet is aligned with the x-axis, i.e. at 6 = 0.
The stator has 12 slots and a three-phase winding. Thus there are two slots
per pole per phase. Each phase winding consists of two adjacent full-pitch coils
of N, turns each, whose axes are displaced from one another by 30 degrees.
The winding is a single-layer winding, and any slot contains A^ conductors
from only one phase winding. This winding is equivalent, in the active length,
to a degenerate concentric winding with only one coil per pole per phase,
having a fractional pitch of 5/6. This is a more practical winding than the one
analysed because it has less bulky endwindings and is generally easier to
assemble. For the same reason, its copper losses are lower.
Consider the flux-linkage tpl of coil a 1 A 1 as the rotor rotates. This is shown
in Fig. 4.5(c). Note that 6 now represents the movement of the rotor from the
reference position in Fig. 4.5(a). The flux-linkage varies linearly with rotor
position because the airgap flux-density set up by the magnet is constant over
each pole-pitch of the rotor. Maximum positive flux-linkage occurs at 0 and
maximum negative flux-linkage at 180°. By integrating the flux-density
around the airgap, the maximum flux-linkage of the coil can be found as

and the variation with 0 as the rotor rotates from 0 to 180° is given by

<M0) = 1 •Ai,
njl

The e.m.f. induced in coil a[A t is given by

_d>p1_ dt// 1 df?_ di/f,
Ci (a
~ di ~ ~ ~d0
which gives
c 1 = 2 V 1 5 g / r 1 c j V.
This represents the magnitude of the square-wave e.m.f. e a ] shown in
FIG. 4.5. Brushless d.c. motor with ideal waveforms of flux-density, e.m.f., and
current, (a) Motor showing two coils of one phase, (b) Magnetflux-density around the
airgap. (c) Flux-linkage of coils 1 and 2 as the rotor rotates, (d) e.m.f. waveform of
coil 1. (e) e.m.f. waveform of coil 2. (0 e.m.f. waveform of phase a. (g) Ideal phase
current waveforms, (h) Switching pattern of switches in the converter of Fig. 4.6(a).
Fig. 4.5(d). Note that the waveform of e.m.f. in this full-pitch coil with respect
to time is an exact replica of the flux-density waveform with respect to position
around the rotor in Fig. 4.5(b).
The e.m.f. induced in the second coil of phase A is identical, but retarded in
phase by 30°. This is shown in Fig. 4.5(e). If the two coils are connected in
series, the total phase voltage is the sum of the two separate coil voltages, and
this is shown in Fig. 4.5(f). The basic effect of distributing the winding into two
coils is to produce a stepped e.m.f. waveform. In practice, fringing causes its
corners to be rounded, as shown by the dotted lines. The waveform then has
the 'trapezoidal' shape that is characteristic of the brushless d.c. motor. With
180° magnet arcs and two slots per pole per phase, theflat top of this waveform
is ideally 150° wide, but in practice the fringing field reduces this to a somewhat
smaller value, possibly as little as 120°. The magnitude of the flat-topped phase
e.m.f. is given by

e — 2N phBJr loy'V
where jVph is the number of turns in series per phase. In this case
^VPh = 2N 1

because the two coils considered are assumed to be in series. In a machine with
p pole-pairs, the equation remains valid provided N b is the number of turns in
series per phase and co is in mechanical radians per second.
Figure 4.6(g) shows an ideal rectangular waveform of phase current, in
which the current pulses are 120 electrical degrees wide and of magnitude I.
The positive direction of current is against the e.m.f., that is, positive current is
motoring current. The conduction periods of the three phases are symmetri-
cally phased so as to produce a three-phase set of balanced 120° square waves.
If the phase windings are star-connected, as in Fig. 4.6(a), then at any time
there are just two phases and two transistors conducting.
During any 120° interval of phase current the instantaneous power being
converted from electrical to mechanical is
P = coT e = 2eI.
The '2' in this equation arises from the fact that two phases are conducting.
Using the expression derived above for the e.m.f., the electromagnetic torque
is given by
T c=4N phBglr lI N m.
This equation is valid for any number of pole-pairs. The similarity between the
brushless motor and the commutator motor can now be seen. Writing E— 2*K
to represent the combined e.m.f. of two phases in series, the e.m.f. and; torque
equations can be written in the form
E=kQ>œ and T=k®i
FIG. 4.6. Converter or inverter of brushless d.c. motor, (a) With star-connected phase
windings; (b) with delta-connected phase windings.

where
k=4N ph and $ = Bgr 1nl.

k is the 'armature constant' and 3> is the flux. These equations for e.m.f. and
torque are exactly the same as for the d.c. commutator motor; only the form of
the constant k is different. It is clear that with ideal waveshapes and with
perfect commutation, these equations are true at all instants of time. The
electronic commutation of the converter switches has thus assumed the
function of the mechanical commutator in the commutator motor, to give a
pure 'd.c.' machine with constant, ripple-free torque.
In practice, of course, none of the ideal conditions can be perfectly realized.
The main result of this is to introduce ripple torque, but the basic relationships
of e.m.f. proportional to speed and torque proportional to current remain
unchanged.

4.4 Torque/speed characteristic: performance and efficiency

The torque/speed curve of the ideal brushless motor can be derived from the
foregoing equations. If the commutation is perfect and the current waveforms
are exactly as shown in Fig. 4.5(g), and if the converter is supplied from an
Scanned by CamScanner
Scanned by CamScanner
Scanned by CamScanner
Basic theory of six-step commutation method

SENSING AND SWITCHING LOGIC SCHEMES OF BLDC MOTOR

Stator winding connection of a three-phase BLDC motor

The following figure shows how to generate the magnetic field in the stator. Here, the positive current is defined as the
current flowing into a specific phase, or coming out of a specific phase.

Fig-7 Magnetic field generation

Similar to the DC motor, if the MCU and control circuit in a BLDC motor do not change the direction of the magnetic field
generated by the stator windings in time, the rotor won’t spin. In BLDC motor, a rotating magnetic field should be generated
by the windings. Therefore, there must be a way to conform the position of commutation and change the direction.If the
rotor wants to spin stable clockwise or counterclockwise, an associated rotating magnetic field must be generated from the
stator windings, which will attract or repel the permanent magnetic (rotor).

Production of Rotating magnetic field

As shown in Figure 7, each phase of the stator coil can generate the magnetic field in two directions and so, the current and
the rotating magnetic fields in the three-phase coils can be easily controlled. Six patterns of magnetic fields (see the following
figure) generated are the basis of six-step commutation, which is explained in the following section.
 Basic theory of six-step commutation method

Figure 8. Rotating magnetic field

Basic theory of six-step commutation method

Six-step commutation
The Hall effect sensor is a sensing switch that outputs a logic level based on the magnetic field detected. The Hall effect
sensors (Ha, Hb, and Hc) are inserted into the stator.
For example, when the Ha sensor is under the N pole of the permanent magnet, it will output signal 1, otherwise 0. See the
following figure.
Combining the outputs of all the three sensors will theoretically give 8 status from 000 to 111. However, in most cases,
because of the hardware constraint, signal 000 and 111 don’t exist. So, the other 6 status can divide the one electrical 360° of
position into six areas, and the exact point where the status changes from one to another is the position that the commutator
changes the direction of the stator’s magnetic field.

Figure 9. Hall sensor output

Figure 10 depicts an example of commutation where the Hall sensor status is shown to be 010. Now, for the rotor to spin
clockwise, the clockwise rotating magnetic field must be generated in its nearest area, that is, where the Hall sensor status is
11. This direction of magnetic field can be generated by turning on the coil AC, which means that the current flows into A,
and runs out of C. When the rotor runs to the area of 011, the Hall sensor status changes to 011 and at the same time, the
commutator changes from AC to BC, to keep the rotor running after the rotating magnet.
Thus, the power sequence is AC -> BC -> BA -> CA -> CB -> AB -> AC.
Here is the summary of the commutation process.
• In one complete rotation of 360 electrical degrees, the excitation of the stator windings will be changed six times, and
each change is called a commutation.
• The angle between S-N pole (rotor) and magnet field (stator windings) is 60-120°, commutation happens at 60°.
• The commutation position is when the status of the Hall sensor changes.
• At every moment, only two phases have current, while the third one is powered off.
 Basic theory of six-step commutation method

Figure 10. Commutation

3.1 Commutation table

As discussed in Six-step commutation, the six commutation positions are fixed in a 360 electrical degrees. So, a special table
can be built to describe the relationship between the sensor status and stator winding excitation, which is called commutation
table. With this commutation table, the MCU can easily control the commutation.

Table 1. Commutation table

Hall sensors Phase
a b c A B C
0 1 1 NC + –
0 0 1 – + NC
1 0 1 – NC +
1 0 0 NC – +
1 1 0 + – NC
0 1 0 + NC –

Following is the terminology used in the commutation table.

• Hall sensors header column provides the Hall sensor status captured from the motor.
• Phase header column determines how the stator windings are excited.
• ‘+’ means the current flows into that terminal.
• ‘-’ means the current flows out of that terminal.
• ‘NC’ means no voltage is applied on that terminal.

Applying this table to Figure 10, it is easy to control the motor spin clockwise and counterclockwise.
• If the Hall sensor status is changed to 100, check Table 1 and then let phase CB turn on; the rotation is clockwise.
• Reversing this, if the Hall sensor status is 011, check Table 1 and then let phase BC turn on; the rotation is
counterclockwise.
Scanned by CamScanner
Scanned by CamScanner
Scanned by CamScanner
Scanned by CamScanner
Scanned by CamScanner
Scanned by CamScanner
Scanned by CamScanner
Sensor Based Control of BLDC motors

Scanned by CamScanner
Scanned by CamScanner