World Academy of Science, Engineering and Technology

International Journal of Electrical, Computer, Energetic, Electronic and Communication Engineering Vol:7, No:1, 2013

Emergency Generator Sizing and Motor Starting

Analysis
Mukesh Kumar Kirar, Ganga Agnihotri

International Science Index, Electrical and Computer Engineering Vol:7, No:1, 2013 waset.org/Publication/17299

AbstractThis paper investigates the preliminary sizing of

generator set to design electrical system at the early phase of a
project, dynamic behavior of generator-unit, as well as induction
motors, during start-up of the induction motor drives fed from
emergency generator unit. The information in this paper simplifies
generator set selection and eliminates common errors in selection. It
covers load estimation, step loading capacity test, transient analysis
for the emergency generator set. The dynamic behavior of the
generator-unit, power, power factor, voltage, during Direct-on-Line
start-up of the induction motor drives fed from stand alone gene-set is
also discussed. It is important to ensure that plant generators operate
safely and consistently, power system studies are required at the
planning and conceptual design stage of the project. The most widely
recognized and studied effect of motor starting is the voltage dip that
is experienced throughout an industrial power system as the direct
online result of starting large motors. Generator step loading
capability and transient voltage dip during starting of largest motor is
ensured with the help of Electrical Transient Analyzer Program
(ETAP).

KeywordsSizing, induction motor starting, load estimation,

Transient Analyzer Program (ETAP).
I. INTRODUCTION

T is extremely important to initially select the correct

generator set due to high initial cost of generator sets, the
installation, and other equipment involved [1]. The generator
rating is not necessarily to be just sum of the loads if some
non-linear loads present. Conventionally that may be taken
care by margin factor but to determine a cost effective as well
as project required rating of generator, effective loads on
generator must be calculated properly to avoid generator over
sizing [2], [3].
Dynamic conditions of generator-unit are interesting in case
of smaller, often isolated electrical grids. Induction machine
play a very important role in the industrial applications and a
significant number of induction motors are used at critical
points of on board processes [4]. As compared to a utility bus
feed, an on-site Gene-set is a limited source of horsepower,
from the engine, and kVA from the generator. Thus, a gene-set
must be large enough to start as well as run connected motor
loads. At start, an induction motor draws heavy surge of
current from the power system that in turn causes a dip in
system voltage [5]. Flicker can create a problem to both the
generator and the connected loads. The basic requirement for
studying starting analysis of Induction motor are the starting
Mukesh Kumar Kirar, assistant professor, Electrical Engineering, MANIT
Bhopal, Madhya Pradesh, India (e-mail: [email protected]).
Ganga Agnihotri, professor, Electrical Engineering, MANIT Bhopal,
Madhya Pradesh, India (e-mail: [email protected]).

International Scholarly and Scientific Research & Innovation 7(1) 2013

current of induction motor and the voltage flicker during start

up [6]. During motor starting, the voltage level at the motor
terminals should be maintained at approximately 80 percent of
rated voltage [7]. Factors affecting motor starting include
impedance of power distribution systems, the motor terminal
voltage, inertia of the motor and the load. System impedance
has impact on both starting current and motor terminal
voltage. Its impact on the voltage is much greater than on the
current. The starting current using across-line starting is only
slightly different when the impedance of the circuit is
changed, but the voltage drop on the cable changes
significantly [8].
II. SYSTEM DESCRIPTION
A single line diagram (SLD) of a power distribution system
of a large Oil Storage Terminal (OST) is shown in Fig. 1. To
improve system reliability and power quality, the ICP has Grid
connectivity with public power company (PPC) at two points.
PPC power is available at 66 kV voltage level through utility
ties 1 and 2. Grid transformers TR-301 and TR-302 step down
the voltage from the 66 kV level to the 6.9 kV level and
connected to Switchgear SG-201A and SG-201B through
cable respectively. The power is primarily distributed at a 6.6
kV level from two main receiving Switchgear panel SG-201A
and SG-201B. Emergency load is connected to MCC-401 and
MCC-402 and SG-402A and SG-402B panel. Generator rating
is calculated according to the load on the buses MCC-401 and
MCC-402 and SG-402A and SG-402B. When the plant
becomes isolated, due to a utility service outage generator will
supply power to emergency load. Emergency power is
generated at 6.6kV and supplied to load through generator
transformer TR-405 of 6.6/0.415kV.
III. STEPS FOR GENERATOR SIZING
Step 1: Segregation of Loads
It is important to gather a reasonably accurate load schedule
as soon as possible for determining generator rating as well as
overall system design. The detailed loads equipment
information is not available at early stage in project. The
primary sizing calculation have to be made based on estimates
and assumptions and the same must be iterated as more
accurate information becomes available progressively. After
accumulation of load list it is necessary to segregate the loads
in different application categories as listed below:
Category-1: Linear loads like lighting, heating etc. in kW
(KWl) and its overall pf (PFl)
Category-2: Running highest motor or a group of motors

71

scholar.waset.org/1999.5/17299

World Academy of Science, Engineering and Technology

International Journal of Electrical, Computer, Energetic, Electronic and Communication Engineering Vol:7, No:1, 2013

like pf (PF
Fs), starting cuurrent (Is), staarting pf (PF
Fss) and
s
starting tim
me (t ) based onn load type.

International Science Index, Electrical and Computer Engineering Vol:7, No:1, 2013 waset.org/Publication/17299

to be starteed at a time or
o sum of autto acceleratedd motor
loads inputt power in kW
W (KWs) and its other paraameters

Fig. 1 Singgle line diagram

m (SLD) of OST
T

m
loads except
e
VFD and
a soft
Category-3: Running motor
ut power in kW
W (KWm) andd its overall pff (PFm)
started inpu
Category-4: UPS inpuut power in kW (KWu) and its
( u)
overall pf (PF
Category-5: Soft startedd motor inputt power ratingg in kW
(KWf) and its overall pf (PFf)

Step-2: Effecctive Load to G

Generator
It is importannt to note that generator ratiing is not neceessarily
too be just sum of the loads if some non--linear loads present.
p
Conventionallyy that may be taken care byy margin factoor but to
deetermine a cosst effective ass well as projeect required raating of
geenerator, effeective loads oon generator must be callculated
prroperly to avvoid generatoor over sizingg. Effective load
l
to
geenerator to bee calculated ass below basedd on load segrregation
doone in step-1
kW

kW
W

kW

kW

kW

kW

kW

(1)

Nu = 1.25 * 1.4 for 1 phasse and 6 pulse UPS and 1.25* 1.15
foor 6 pulse UPS
S with input fiilters or 12 puulse UPS. Facttor 1.25

International Scholarly and Scientific Research & Innovation 7(1) 2013

is considered too take the batttery charging after drain ouut with
suppplying conneected loads.
Nf = 1.15 for without bypass contactor after
a
start and 1.0 for
wiith bypass conntactor after sttart
Effective KV
VA rating of thhe generator iss

kVA

W
PF
F

(2)

Step-3: Overlload Capacityy

Overloading has to be cheecked with reespect to shorrt time
kV
VA rating not the short time kW rating. During startinng, kW
is low due to loow starting power factor whereas
w
line current
c
still continues to
t 5-7 times oof full load cuurrent up to 80%
8
of
ratted motor sppeed even thoough power factor increaased to
nearly 0.7 to 0.8.
0 In case of
o firewater pump conneccted to
generator, voltaage dip of 155% only to bee considered during
staarting as per NEC.
N
Even if firewater pu
ump starter iss fitted
wiith reduced voltage
v
startinng, DOL to be considerred for
generator sizingg keeping in viiew criticalityy of fire water pump.
Considering starting
s
load, short time tootal kVA loadd rating
to be calculateed as below without con
nsidering corrrection

72

scholar.waset.org/1999.5/17299

World Academy of Science, Engineering and Technology

International Journal of Electrical, Computer, Energetic, Electronic and Communication Engineering Vol:7, No:1, 2013

factors for non-linear loads.

kVA

W
PF

PF

like cables and running loads.

W
PF

W
PF

PF

PF

(3)

where Kr is to be selected from IEEE Std 399-1997

kVA

kVA

(4)

However, if no data is available on overload capacity

= 1.5 could be considered and suitability to be

K
checked.
If above relationship is not fulfilled,

International Science Index, Electrical and Computer Engineering Vol:7, No:1, 2013 waset.org/Publication/17299

kVA correctedinStep

VA

W I K PF
PF

kVA

P. F. ofGenerator

StepLodingFactor (6)

X"

KVA PF /

X"

KVA PF /

(8)

W I K

X"
W I K

(9)

KW

I
PF

X"

100
V

Step-6: Generator Rating

Generator electrical kW kW calculated
kW = Generator kVA kVA *0.8(Gene p.f.)

(10)

Engine i.e. mechanical kW (kWm) selected

kWm = kWe selected / efficiency of engine

(11)

Suitable design / future margin, normally 10 to 30%, should

be applied on above calculated kVAe rating and round off to
nearest rating available in market.
After exercising all of above five steps, Generator rating
shall be maximum rating out of calculation from step-2 to
step-5. According to different application categories, the load
is segregated in Table I. Flow chart for generator sizing is
given in Fig. 2. The generator rating selected after the
calculation is 5MVA. The generator parameters which are
considered in the study are depicted in Table V of appendix.

If it is not fulfilled,
kVA correctedinStep4

Step-4: Step Loading Capacity

Step loading has to be checked with respect to short time
kW rating not with respect to the short time kVA rating. The
effect of step load is much more on engine/governor which is
responsible to deliver the desired kW. Step loading capacity is
fully depends on engine type and associated governor
type/response time. Usually governor type is unknown during
initial phase of project and conservatively maximum 40%
engine capacity could be considered as a permissible step
load. Again maximum starting load in kW shall be selected as
maximum step load. So the relationship given below is to be
checked considering 40% allowable step load. In case accurate
step load factor is obtained from vendor, the same is to be
applied instead of 0.4.

where V is allowable voltage dip

If above relationship is not fulfilled
kVA Corrected in step-5

(5)

X"

%V

TABLE I
LOAD LIST
W I K PF
PF

P.F.

Type of the Load

Lighting/Heating Load (kWl)
Highest Motor Load (DOL) (kWs)
Running Motor Load (kWm)
UPS Load (12 Pulse) (kWu)
Soft Started Motor Load (kWf)
Miscellaneous kW
Total

(7)

Step-5: Transient Voltage Dip

The instantaneous transient voltage dip occurs during
starting of motor, is strictly a function of the relative
impedances of the generator and motor. Gen-set's ability to
start large motors without excessive voltage and frequency dip
is a function of the complete system.
Excitation systems that responds too quickly or too stiff can
actually overload the engine when starting large motors. In
general generator has to recover voltage at least up to 90% of
rated voltage utmost by few cycles. If the motor does not
reach near rated operating speed after few seconds (say 4 to 10
sec), a heavy voltage and frequency dip may occur. It is to be
noted that manual calculation of transient voltage drop does
not lead to any conclusive result. The exciter type, AVR
voltage change rate, and time constants of exciter are major
factors to determine whether generator is capable to recover to
90% voltage after transient dip. Keeping in mind this aspect,
transient voltage dip at motor starting could be calculated by
very simplified method ignoring other parameters of network

International Scholarly and Scientific Research & Innovation 7(1) 2013

73

kW
1000
110
74.8
4
55
2650
3893

scholar.waset.org/1999.5/17299

P.F.
0.9
0.85
0.86
0.8
0.85
0.85

kW/Pf
1111
129.4
86
5
64.7
2887
4515

World Academy of Science, Engineering and Technology

International Journal of Electrical, Computer, Energetic, Electronic and Communication Engineering Vol:7, No:1, 2013

International Science Index, Electrical and Computer Engineering Vol:7, No:1, 2013 waset.org/Publication/17299

induction motor parameters considered in study are depicted

in Table IV of appendix.

Fig. 2 Flow chart for generator sizing

IV. MOTOR STARTING ANALYSIS

To investigate the transient performance of Gene-set during
motor starting, the mathematical models of exciter, governor,
and load is considered in this study as given in appendix. The
transient and subtransient reactances of the generator and
motor are considered in the computer simulation so that more
accurate results can be obtained.
The gene-sets ability to start large motors without large
voltage and frequency dips depends on the entire system.
System factors that affect generator operating characteristic
during motor starting include:
Available engine power;
Capacity of the generator;
Energy stored in the rotating inertia of the gene-set;
Impedance of power distribution systems;
Electrical design parameters of the motor;
Acceleration of the motor and its load (motor
characteristics);
Motor terminal voltage.
A properly sized generator supports the high starting kVA
(kVA ) required and sustain adequate output voltage for the
motor so it can produce the required torque to accelerate its
load to rated speed. After the initial voltage dip, it is important
that the generator restore voltage to at least 90% to develop
adequate torque to accelerate its load to rated speed [9]. The

International Scholarly and Scientific Research & Innovation 7(1) 2013

A. DOL Starting of IM on Gene-Set: Case Study

Emergency loads are usually served by on-side generation
of limited capacity, which generally magnify voltage drop
problem on motor starting, especially when large motors are
involved. The DOL motor starting analysis is performed to
investigate generator performance, induction motor
performance and the performance of other connected load.
The motor starting study can expose and identify the extent of
a voltage drop problem [10]. The case study is performed with
the help of ETAP software.
The case study is executed for 0.415kV motor control
centre (MCC-402) in OST power distribution system. The fire
water pump motor rated at 110kW is connected to MCC-402
as shown in SLD Fig. 1. Motor starting analysis is done when
the entire load supplied by the gene-set is already in operation.
The summary of the load flow results are given in the Table II.
Design parameters of induction motor are given in
appendix. Motor starting current and the moment of inertia of
motor and load is taken according to motor data sheet. The
110kW motor (PM-026A) is connected to MCC-402 by 400m
long cable. The simulation results during DOL starting of
110kW induction motor, starting current, MCC-402 and motor
terminal voltage, slip, motor and load torque are shown in Figs
3-6 respectively.
Voltage difference between voltage at motor terminal and
the MCC-402 is due to voltage drop in the cable. Allowable
voltage drop in the cable during normal operation is 3% and
15% during starting time. The motor terminal voltage during
starting is 84.07% and then recovers slowly and stabilized at
the post starting value of 93.39%. The percentage is based on
nominal bus voltage.
Available Voltage at different buses as percentage of bus
nominal kV and time duration throughout DOL motor starting
is shown in Table III

Bus ID
EPCC-401
MCC-401
MCC-402
PMCC-402A
MPCC-402B
SG-403
PM-026A

Device ID
PM-026B
EPCC-401
MCC-401
MCC-402
PMCC-402A
PMCC-402B
SG-403

74

TABLE II
LOAD FLOW RESULTS
Nominal kV
Operating kV (in %)
0.415
96.87
0.415
96.5
0.415
95.81
0.415
95.73
0.415
95.68
0.415
95.81
0.415
94.54

TABLE III
VOLTAGE DIP DURING MOTOR STARTING
Rated voltage
% Voltage Operation
0.415
84.2
0.415
93.3
0.415
92.8
0.415
90.5
0.415
92.2
0.415
92.2
0.415
92.3

scholar.waset.org/1999.5/17299

Duration (sec)
11.18
11.02
11.16
11.26
11.24
11.24
11.22

World Academy of Science, Engineering and Technology

International Journal of Electrical, Computer, Energetic, Electronic and Communication Engineering Vol:7, No:1, 2013

B. Generator Power and PF

F during DOL
L IM starting
The power drrawn by the iinduction mottor during starrting is
maainly reactivee due to the low
l
power faactor at the time of
staarting. Powerr output of tthe generatorr during the motor
staarting is shown
n Fig. 7.

International Science Index, Electrical and Computer Engineering Vol:7, No:1, 2013 waset.org/Publication/17299

Fig. 3 Motor starting current variatioon with time

Fig. 7 Generaator output pow

wer variation durring motor startting

Fig.
F 4 Motor terrminal and MC
CC-402 voltage in % wrt bus noominal
vvoltage

During motorr starting, low

wer power facctor demands higher
excitation curreents and resullts in increaseed losses. Gen
nerator
uring motor sstarting is sho
own in Fig. 8.
8 Over
poower factor du
sizzing A.C. gen
nerators for operation
o
at lower power factors
ressults in lower operating effiiciency and hiigher costs.

Fig. 8 Generaator power factoor variation durring motor startting

F 5 Motor slipp during motor starting

Fig
s

V. CON
NCLUSION

Fig. 6 Motor
M
and Load torques during motor starting

International Scholarly and Scientific Research & Innovation 7(1) 2013

The dynamic behavior of ggene-set unit during

d
direct-o
on-line
staarting of indu
uction motors is interestin
ng in grid issolated
sysstem. It is concluded from
m above resu
ults the genee-set is
cap
pable of supplying both steady state and transientt loads
wiithout exceed
ding its designed frame ov
verload capaccity or
pro
oducing a dan
ngerously deppressed voltag
ge. Generator sizing
typ
pically depend
ds on the perm
missible percen
ntage of voltaage and
freequency dip and
a the accepttable duration
n of the voltaage and
freequency dip reecovery time. Generator ratting also depeends on
thee percent of load step andd type of loaad connected to the
generator. Based
d on the givenn results it is clear
c
generato
or gives
goood starting performance in terms of lowest voltage
v
flu
uctuation. Acccordingly, it iss concluded frrom results ob
btained
byy motor startiing analysis tthat the moto
or can DOL started

75

scholar.waset.org/1999.5/17299

World Academy of Science, Engineering and Technology

International Journal of Electrical, Computer, Energetic, Electronic and Communication Engineering Vol:7, No:1, 2013

satisfactorily. This paper can be used as a design guide for

generator selection. Motor starting study on gene-set must be
performed at the design stage. If the system shows motor
starting difficulty due to excessive voltage drop in the system,
a new system configuration should be chosen to avoid the
problem. Considering both the cost and performance of a
standalone gene-set system should be carefully evaluated to
select appropriate size during the design stage.

APPENDIX
A. IEEE Type 1 - Continuously Acting Regulator and Exciter

International Science Index, Electrical and Computer Engineering Vol:7, No:1, 2013 waset.org/Publication/17299

Fig. 9 Exciter block diagram

kW

kV

110

0.415

Full Load
Current
179

P.F.
0.88

TABLE IV
INDUCTION MOTOR DATA
LRC in % of
Starting P.F.
LRT in % of full load
full load current
torque
600
0.25
98.69

Max. Torque in %
of full load torque
199

X/R

11.1

0.92

TABLE V
GENERATOR DATA
kVA
5000

kV
6.6

P.F.
0.85

H
2.4

"

"

0.11

0.12

0.23

REFERENCE
[1]

IEEE Recommended Practice for Industrial and Commercial Power

Systems Analysis, IEEE Std 399-1997 , vol., no., pp.I, 1998
[2] IEC 60034-12:1980, Rotating electrical machines Part 12: Starting
performance of single speed three-phase cage induction motors for
voltages up to and including 660 V
[3] Rotating electrical machines Part 1: Rating and performance, IEC
60034-1, 1999-08.
[4] IEEE Recommended Practice for Protection and Coordination of
Industrial and Commercial Power Systems," ANSI/IEEE Std 242-1986 ,
vol., no., pp.0_1, 1986.
[5] A. Jack Williams, M. Shan Griffith, "Evaluating the Effects of Motor
Starting on Industrial and Commercial Power Systems," Industry
Applications, IEEE Transactions on, vol.IA-14, no.4, pp.292-305, July
1978.
[6] P.S. Patil, K.B. Porate, "Starting Analysis of Induction Motor: A
Computer Simulation by ETAP Power Station," Emerging Trends in
Engineering and Technology (ICETET), 2009 2nd International
Conference on , vol., no., pp.494-499, 16-18 Dec. 2009
[7] IEEE Recommended Practice for Electric Power Distribution for
Industrial Plants IEEE Std 141-1993.
[8] Liang Xiaodong, O. Ilochonwu, "Induction Motor Starting in Practical
Industrial Applications," Industry Applications, IEEE Transactions on ,
vol.47, no.1, pp.271-280, Jan.-Feb. 2011
[9] John H. Stout Capacitor Starting Of Large Motors IEEE Transactions
On Industry Applications, Vol. Ia-14, No. 3, May/June 1978, pp. 209212
[10] A. Jack Williams, JR. Member, M. Shan Griffith, Evaluating the
Effects of Motor Starting on Industrial and Commercial Power Systems
IEEE Transactions On Industry Applications, Vol. IA-14, NO. 4,
JULY/AUGUST 1978

1.1

0.12

1.08

0.002

5.6

0.002

respectively. She is currently working as a professor in the Department of

Electrical Engineering, MANIT, Bhopal, India. She has 12 research papers in
International journals, 20 research papers in National journals, 22 research
papers in International Conferences and 70 research papers in National
Conferences. Her fields of interest are Power System Planning, Power
Transmission Pricing, Power System Analysis and Deregulation. Dr.
Agnihotri has a membership of Fellow IE (I) and LISTE.

Mukesh Kumar Kirar was born in Narsinghpur, India, in 06 Feb 1983. He

received the B.E. (Electrical) degree from Government Engg. College, Ujjain,
India in 2006 and M.Tech. (Power System) in 2008 and pursuing Ph.D from
MANIT Bhopal, India. He is currently working as an assistant professor in the
Department of Electrical Engineering, MANIT, Bhopal, India. His field of
interests are power system stability and control, transformers and machines.
Ganga Agnihotri was born in Sagar, India, in 27 May 1949. She received the
B.E. (Electrical) degree from MACT, Bhopal, India. She received the M.E.
(Advance Electrical Machine) and PhD (Power System Planning Operation
and Control) from University Of Roorkee, Roorkee in 1974 and 1989

International Scholarly and Scientific Research & Innovation 7(1) 2013

"

"

0.15

76

scholar.waset.org/1999.5/17299