Unit 2 Main Dimensions of A.C.

Machines
2.1 Main dimensions

Problem: to determine the air gap diameter D and core length L for a
specified speed and apparent power VA (voltage×current).

is Bg = flux density when no

slotting effect? p.29
D

For convenience in design, the following specific loadings are

defined.
1
1. Specific magnetic loading B

B is defined as: the average magnetic flux density over the whole
surface of the air gap.
The surface area of the air gap in one pole:
D entire premeter * 1/total pole * L = surface area of one pole pair
a= L
2p (The perimeter is: D. The total surface area
of the air gap is: D×L.)
where p = number of pole pairs; L = axial length of iron core.
B=  m
  m

2p  m
D
a L DL T (Wb/m2) (1)
2p

where Φm = maximum flux per pole.    m sin(2ft )

Because of the saturation in iron core, B has its limitation value.

2
Note that B directly affects:

 core loss  B2
 magnetizing current (hence power factor) in induction motors
why???
B  magnetizing current  power factor

Typical values of B : 0.44 - 0.64 T for synchronous machines

(in the air gap) 0.3 - 0.48 T for induction machines

pf lower because pf lower because

magnetizing current When there is a large magnetizing current, it means
increases -> Q =I^2 *R ;
since Q larger, angle that a significant amount of reactive power being
larger, so pf smaller drawn.
since Pimg larger

3
2. Specific electric loading ac

ac is defined as (the number of conductors)×(the rms value of current

in the conductor) per unit length of gap-surface circumference, or the
rms peripheral current density.
NI rms
ac 
l
where N: number of all conductors

Irms: root mean square current I rms 

1 T
 i t 2
dt
T 0
l: length of gap-surface circumference
The number of all conductors along the air gap:
3 2Tph (for 3 phase machine)
m  2Tph (for m phase machine)

where Tph = turns number per phase. Therefore,

2* Tph = conductor per phase 4
 m  (2T ph I ph ) total rms current in the
ac  conductors per unit lenght
D (2)
where Iph = rms current per phase.

Note that ac  affects:

 copper losses 
 armature heating 
 the voltage drop on leakage reactance 
 the armature reaction in synchronous machines  armature reaction is
proportional to phase
current
Because of the concern of temperature rise, ac has its limitation value.

In small machines ac varies between 8000 to 25000

ampere-conductors/m.

5
 (in tutorial , q.2)i
3. Output coefficient G angular velocity, w= 2pi*n, not 2pi*f
angular frequency, w= 2pi*f

The number of turns per phase: Tph

The number of effective turns per phase: KwTph
The winding factor: Kw (coil-span factor  distribution factor) Kp x Kd
The maximum flux per pole: Φm , where    m sin(2ft )
The maximum flux linkage per pole: KwTphΦm

The electromotive force in one phase:

e 

d d K wTph m sin(2ft) 
 2fKwTph m cos(2ft)
dt dt
The rms electromotive force in one phase:
E
emax 2fKwTph m
  2fKwTph m  4.44K wTph m f
2 2
E  4.44KwTphm f is it the same as previous course???
The voltage drop of one phase is approximately equal to the E:
6
 Vph  E  4.44KwTphm f

(Apparent power VA)2 = (Real power)2 + (Reactive power)2

The VA of an a.c. machine is given by the equation:

S  3Vph I ph  3  (4.44KwTphm f )I ph Becuase V ph  E  4.44KwTphm f 
2p m DLB
Because B=
 m =
DL 2p
f = np (n: r/s revolutions/second)
3  (2Tph I ph ) ac  D
ac  I ph  ac = electric loading
D  3  2  Tph
  DL B   ac  D
S  3   4.44K wT ph     (np)  

  3 2 T
  2 p   ph

 1.11Kw 2 B  ac  D2 Ln (3)
S is VA for input
The output coefficient G is defined as
7
 S defined as
 1.11K w B  ac  G
2 S = apparent
2
D Ln power (4)
G  1.11K w 2 B  ac is almost a constant.
Note that for a given power or VA, the volume of the machine is
inversely proportional to the speed. Thus for efficient use of materials,
use high-speed machines.

If the output power and the speed are given, the volume of the
machine is almost determined:
S n = revolution per phase
D L
2

Gn
Therefore, when we design an electric machine, we can first
determine the volume D2L.

8
 4. Ratio of L to D (or ) :tau

 For induction motors running at normal speeds, a roughly square

pole will usually give satisfactory design.
pole-pitch = curmanstance/pole number
 D 
 
 Square pole: pole-pitch τ  2 p  is equal to the core length L i.e.
 L (5)
(Total circumference of air gap: D)
In general:
L L
  0.6 ~ 2.0
 D
(5a)
2p
If L is longer,  end winding becomes smaller
since winding smaller,
less resistance
 copper loss becomes smaller.

If L is too long, is good or not?

9
 L L
  0.6 ~ 2.0 
 0.6 ~ 2.0
L
 D
 D 2p
2p
For high-speed synchronous generators:
 high radial centrifugal force (離心力)
 the hoop stress caused
 D is limited (D rarely exceeds 1.2 m for 2-pole machines)
 need to use large L/D ratios e.g. 4:1

For low speed synchronous generators: if n smaller, p bigger

pn
f 
 the large number of salient poles, because 60 , f=constant
 large D ratio required in order to accommodate the large number of
poles
 small L/D ratio, say 1:5 (Fig. 3) D: air gap diameter

10
(a) Rotor of turbogenerator (b) Rotor of hydro-turbine generator
(High speed, large L/D ratio) (Low speed, small L/D ratio)
Fig. 3 Main dimensions of rotating machines

Turbogenerator Hydro-turbine generator

(steam powered) (moving-water powered)

11
2.2 Magnetic circuits of electric machines

The basic relationship governing magnetic circuits is

Magnetic circuit is analogical with Electric circuit
F=R×Φ V=R×I
m.m.f. = Reluctance × Flux
where R = l/(μa). 4 x pi × 10-7 m kg s-2 A-2 R = l/(a) ρ=σ, sigma =1/ρ
resistivity
where a: the area, µ: permeability (導磁率)
Alternatively,
Φ=×F I=G×V
 : Permeance (磁導) G: Conductance
where  is the reciprocal of R :
 = 1/R = (μa)/l. G = 1/ R = (a)/l
L = inductance
The connection between magnetic circuit and electric circuit:
Li = NΦ (NΦ is flux linkage)
therefore L = NΦ/i = N(F)/i = N(Ni)/i = N2. (F = Ni)
12
One important concept: for linear materials (µ is a constant),
inductance only depends on the structure, material, and the number
of conductors. It doesn’t depend on any sources.

For magnetic circuit design: if we know Φ, how to calculate F

(1) Divide the magnetic circuit into sections of uniform cross-sections.

(2) Assuming the flux density to be the same everywhere over each
cross-section of the same section and given by Bi = Φ/ai. (ai is the
area)

(3) For linear materials, Hi = Bi/μi. H: the magnetic field intensity

(magnetic field)
For nonlinear materials, according to B-H curve, get Hi

(4) Then for the ith section, the m.m.f. drop is Fi = Hili.
13
(5) If all sections are in series, we have
Fi = H1l1 + H2l2 + ...+ Hnln = Ni = F
 source

Example of magnetic circuit calculation

14
uniform for
each area

15
B-H curve modelling*
 The B-H curve is required for calculating the m.m.f. of certain
sections if magnetic saturation is present.
 For manual design calculations, the B-H curves may be consulted
directly.
 For computer-aided design, it is necessary to model the B-H curve
using one of the following methods:

(1) Piecewise linearization method: the B-H curve is divided into

sections and each section is described by a straight line segment
(Fig. 5). However, some programming effort is needed as the
computer needs to check which section is to be used in the
calculation.

16
 linear region

Fig. 5 Piecewise linearization of B-H curve

(2) Mathematical function: the B-H curve is represented by a suitable

mathematical function, which can be algebraic, exponential, or a
power series. For example,

B = aH/(1 + bH), where a and b are constants, or

B = (a0 + a1H + a2H2 + ...)/(b0 + b1H + b2H2 + ...)

17
 (Rational function)
B  aHb (Geometric function)
N
B  a0   exp(bn H )
n 1
(Exponential function)
N
H   an Bn
n 1
(Power series)

Working flux densities B

The maximum flux densities normally used:

Core plates. Transformers (c.r.o.s.): 1.3 - 1.6 T

(cold-rolled grain-oriented steel strip)
Machines: Teeth: 1.8 - 2.2 T
Other regions in cores: 0.8 - 1.2 T
Cast steel. Poles: 2.0 T
Yokes: 1.5 T
18
Electric machine magnetic circuits
Stator yoke
Stator pole
Air gap
Rotor teeth
Rotor yoke

The magnetic circuit of a rotating machine consists of:

1. Stator yoke
2. Stator teeth (or stator pole)
3. Air gap
4. Rotor teeth (or rotor pole)
5. Rotor yoke
19
 Stator yoke f/2
Stator teeth f
yoke teeth
Air gap f
stator x/2 x
rotor kx/2 kx Rotor teeth kf

Rotor yoke kf/2

Fig. 6 Magnetic circuit of a rotating machine

(Here suppose the source is in rotor, k > 1)
 Note that if the main flux crossing the air gap is Φ, then the flux in
the stator yoke is Φ/2.
 Due to flux leakage, the flux in the rotor teeth is kΦ, where k is the
leakage coefficient. flux in air gap = useful flux, phi,g
leakage flux = phi,l 20
leakage coefficient = (phi,g+phi,l)/phi,g
Flux leakage:

(If the source is in stator, k < 1.)

flux density in stator’s yoke
Byoke

the flux density

Bpole _ head
in the pole head

flux density
Bpole _body
in the pole body

21
2.3 Magnetic circuit calculation in air gap

1. Slotting effect
Slotting effect:

if air gap is uniform, flux is uniform.

However, since there are slots, so
flux is not uniform

Given Φ (flux per pole), how to calculate the m.m.f. drop in air gap Fg

22
 pole arc

bt + f.bs
axial length

bt bs

Fig. 7a Slotting effect

We know: Φ = flux per pole, Lc = core length, β = pole arc.

(1) Consider two cases

Case 1 If assuming no slotting effect, then the flux density under the
pole (in the air gap) is no slotting effect, so uniform

  
Bg   
a Lc  surfacearea of teeth  slots (6)
The air gap m.m.f. is
Bg
Fg  H g g  g = mag. field * air gap
0 height (7)
23
Because the presence of slots results in slotting effect (Fig. 7), the flux
lines do not cross the air gap in straight lines.

Because there are slots, the actual area which the flux passing is less
than a  Lc  .
The real flux density will be larger than that

Bg 
calculated from Lc  .
Case 2 Consider another extreme situation:
assuming all the flux to be confined to the teeth area:

Bg  teeth < (teeth + slots);
surfacearea of teeth B' (real flux density)> B

The real flux density Bg will be smaller than the value calculated
under this assumption.
Conclusion
That is, the real Bg :
24
  
 Bg 
surface area of teeth  slots surface area of (teeth)
(2) The method to deal with this problem
One method to account for slotting effect is to
consider the effective tooth width given by
Effective width = bt + fbs (0 < f < 1)
where f = a function of the ratio bs/g.
We need to calculate: Φ  the real Bg . wait, here bt
and bs are
exchanged
Given Φ, consider two methods:
Method 1 Assuming the flux crosses the air gap in straight lines over
the effective width. So the flux through one slot pitch is:
Bg bt  f  bs Lc (from this expression, Bg can be determined)
(At this step, we already solve the problem. slot pitch: distance
between 2 adjacent slots
However, in practice, we would like: Φ  Bg  the real Bg .)

25
Method 2 Assuming that there is no slotting effect. Bg is the calculated
value of flux density under this assumption. The flux through one slot
pitch is:
Bg bt + bs Lc (from this expression, Bg can be determined)
Using the two methods the values of the flux should be the same
(given condition):
Bg bt  f  bs Lc  Bg bt + bs Lc
The real flux density in the air gap is now:
bt + bs
Bg   Bg = K s Bg ideadl Bg greater than the
bt  f  bs Bg
where
bt  bs bt  bs s s
Ks    
bt  f  bs bt  bs  bs  f  bs  s  (1  f )bs  s  ks  bs
s
Ks  Ks  1
 s  ks  bs (8)

26
In (8), s = slot pitch = bt + bs , and ks = Carter’s coefficient for slots
(which is given in Fig. 7b).
larger than theslot,
but smaller than
the whole area

Fig. 7b Carter coefficients for air gaps

(Taken from Ref. [1])

2. Effect of ventilating ducts

Vent ducts are necessary in large machines for better heat dissipation.
Both radial and axial vent ducts may be present (Fig. 8).

27
 Fig. 8 Ventilating ducts in large machines

Radial vent ducts have the same slotting effect as open slots on the air
gap flux density, hence another factor Kd is required:
Lc s
Kd  Ks 
Lc  kd  nwd  (9) (Compare with:  s  ks  bs )
where Lc = overall core length, wd = width of vent ducts, n = number
of vent ducts, kd = Carter’s coefficient for vent ducts (determined from
Fig. 7b).
28
If consider the effects of slotting and vent ducts,

Bg  K s Kd Bg  K g Bg Bg = flux/((bt+bs)*Lc)

where Kg = KsKd. The air gap m.m.f. is now modified to

Bg K g Bg Bg 1 1
Fg  g= g=  Kg g  Bg  g  Fg  Bg  g 
0 0 0 0 0 (10)

The effect of slotting and vent ducts may be treated as if the air gap is
increased to g  K g  g ( Kg  1 ).
we extract Kg from Bg' to make g'

29
3. Fringing effect at the iron core edges and at the pole edges
Flux fringing effect:

When we calculate flux or flux density in the air gap, the flux fringing
effect can be accounted for by modifying the axial length and pole arc
as follows:
Lc  Lc  k1  g(axial length)
     k2  g (pole arc) (11)
 k1 and k2 are constants allowing for increase of effective
cross-sectional area for passage of flux through the air gap.
 In practice, k1 = k2 = 2 gives fairly accurate results for small air
gaps.
30
4. Choice of air gap length in induction motors

 The air gap length is determined by:

 the magnetizing current (too low a power factor should be
avoided)
 mechanical considerations (bearings, unbalanced magnetic pull
etc.)
 cooling condition

 As a rule the air gap should be made as small as possible as is

mechanically possible.

 In small motors it may be of the order of 0.25 mm. A typical

expression for estimating the gap length is
g  0.2  2 DL mm (12)
where D and L are in m.

31
2.4 Frames of electric machines

What’s the meaning of D71-20 on the nameplate of a motor?

D: frame is totally enclosed
71: frame size (= centre height)
20: shaft diameter

 A frame is the mechanical structure

housing a stator of a given outside
diameter and the auxiliary components.
 Standard series of frames are used for different power
ratings.
 Standard frame sizes prescribed by international standards,
e.g. IEC 72 1971 and its amendments in 1977 and 1981.

32
C centre heights L - shaft diameter
D M
E S

Foot-mounted machines

Flange-mounted machines

33
1. Dimensions of machines:

 For foot-mounted machines, the machine’s centre height (中心高),

expressed in minimeters, designates the frame size, or the frame
number.

 the centre height

 When more than one core length is used, the frame number will
consist of the machine’s centre height followed by a letter
indicating the core length:
type of closure + frame number + - + shaft
 letter S for short core diameter (in mm)
 M for medium core frame number= center height (in mm)
if more than 1 core length is used, use
 L for long core center height + S/M/L

34
 By choosing different core lengths, two or more ratings can be
provided by the same frame.
For example, 90L: 90mm shaft centre height, and long core
frames 90L and 90S: both have a shaft centre height of 90 mm
frame 90L: has power rating of 1.5 kW;
90S: has power rating of 1.1 kW.

 A letter appearing before the frame number indicates the type of

enclosure:
 C for enclosed ventilation
 D for totally enclosed
Totally-enclosed machine systems block the
free exchange of air between the external
environment and the components within the enclosure. A total
enclosure provides complete coverage of system devices, but is
not airtight or hermetically sealed.

35
  E for flameproof

 A dash is used to separate the frame number and the shaft diameter
if the frame number does not end with a letter (the core length S,
M, or L).
For example, D71-20 denotes a totally enclosed machine (D),
whose centre height is 71 mm and shaft diameter is 20 mm.

2. Dimensional standards: (dimensions of mounting holes)

IEC Publication No. 72 gives the full details of standardized dimensions of
foot-mounted and flange-mounted machines. With reference to Fig. 9, the
designations of the main fixing dimensions of a foot-mounted machine are:
 H: standardized centre height;
 A, B: distance between centre lines of mounting holes from end
view and side view, respectively;
 C: distance from shaft shoulder to centre line of mounting holes
at driving end.
36
Fig. 9

37
Fig. 10 gives the main standardized fixing dimensions for
foot-mounted machines for frame sizes 56 to 200

Fig. 10

38
3. Standardized ratings:

The IEC recommended ratings in kW for small machines as:

0.06, 0.09, 0.12, 0.18, 0.25, 0.37, 0.55, 0.75, 1.1, 1.5, 2.2, 3.7, 5.5, 7.5,
11, 15, 18.5, 22, 30, 37.

These twenty different ratings can be provided with only 11 standard

frames.

Fig. 11 gives the technical details of frame sizes for standard four-pole
motors.

39
 Fig. 11

(Fig. 9, Fig. 10 and Fig. 11 are taken from Ref. [3].)

40
References:

[1] ‘Alternating Current Machines’ by M.G. Say, 5th ed., ELBS

[2] ‘Electrical Machinery’ by A. Draper, ELBS

[3] ‘Design of Small Electrical Machines’ by E.S. Hamdi, Wiley

41