0 IEEE Transactions on Energy Conversion, Vol. 10, No.
1, March 1995
SELECTION OF CAPACITORS FOR THE SELF REGULATED SHORT SHUNT
L. Shridhar, Student Member, Bhim Singh, C. S. Jha, B. P. Singh, SM and S. S. Murthy, SM
Department of Electrical Engineering, IIT-Delhi
Hauz Khas, New Delhi-110 016 INDIA
Abstract: The self regulating feature of a
Self Excited Induction Generator (SEIG) by
connecting additional capacitors is examined.
The system consisting of both shunt and series
capacitors has been analyzed. A methodology
has been explained to choose appropriate set
of values of these capacitors for desired
voltage regulation. Performance of short and
long shunt configurations has been compared
which shows the superiority of the short shunt
connection. Results of a detailed
investigation on a conventional 3.7 kW
induction motor operated as a SEIG are
presented to illustrate the effectiveness of
the proposed method. Close agreement between
predicted and test results has been observed
thereby establishing the validity of the
analysis carried out and the criterion
adopted.
Key Words: Induction Generator, Capacitor Self
Excitation, Autonomous Power Generation
INTRODUCTION
Capacitor self excitation of induction
machine is now a well known phenomenon which
has been researched in depth [1-18]. If an
appropriate capacitor bank is connected
across the terminals of an externally driven!
induction machine, a voltage is developed
across its terminals. The residual magnetism
in the rotor initiates voltage build-up which
is augmented by the capacitor current to cause
a continuous rise in voltage. A steady state
voltage results due to the magnetic
saturation which balances the capacitor and
the machine voltage.
Over the years, the authors have been
engaged in investigation related to analysis,
design and control aspects of SEIG with a view
94 WM 117-2 EC A paper recommended and approved
by the IEEE Electric Machinery Committee of the
IEEE Power Engineering Society for presentation
at the IEEE/PES 1994 Winter Meeting, New York,
January 30 - February 3, 1994. Manuscript sub-
mitted August 10, 1992; made available for
printing December 13, 1993.
SELF EXCITED INDUCTION GENERATOR
to evolve viable standby/autonomous power
generating units driven by oil engines, micro
hydro turbines and wind turbines [7,9,12,13,15
-18]. Brushless rotor construction, lower unit
cost, absence of a separate dc source, better
stability and self protection under fault
conditions are the major reasons for
preferring SEIG over conventional alternator
in such generating units. But poor voltage
regulation of SEIG even at regulated speeds
has been a major bottleneck in its
application. Steady increase in capacitor VAR
with load has to be achieved to maintain good
voltage regulation [1-8]. Several voltage
regulating schemes have been tried to achieve
this aim [3-6,9,10,13,16,18]. These schemes
mostly utilize switched capacitor or variable
inductor or saturable core reactor based close
loop schemes using relay/contactors or
semiconductor switches. But complex system
configuration, intricate control circuit
design and operational problems like harmonics
and switching transients, associated with
voltage regulators vitiate the very advantages
of recommending induction machines for
autonomous power generation.
Inclusion of additional series capacitance
to provide additional VAR with load is one of
the attractive options to improve voltage
regulation of SEIG. Basset and Potter [1] in
their classical paper have presented some
experimental results to highlight the self
regulating feature of compounded SEIG. It is
also demonstrated experimentally that it is
possible to have almost a flat load
characteristic with a particular combination
of capacitors connected under short shunt
configuration (Figure 1 ) . This eliminates any
additional voltage regulator since the scheme
has inherent self regulating feature. But the
paper is limited to qualitative explanation
and some experimental results as it does not
Capacitance
Fig. 1 Block diagram of short shunt SEIG system
 11
throw light on analytical methods and design
tools. In a recent paper, Bim etal [14]
reported an approximate method of analyzing
compounded SEIG under long shunt
configuration.
Based on the work reported so far, it was
felt necessary not only to develop analytical
techniques to predict the performance of short
Fig.2(a) Equivalent circuit of short shunt SEIG
shunt SEIG, but also to evolve a methodology
to select the most appropriate pair of Xce = per phase capacitive reactance of the
capacitors. An ideal combination of shunt and series capacitor, C s e
series capacitors is the one for which the RL = per phase load resistance
load voltage can be maintained within F,V = p. u. frequency and speed
acceptable limits from no load to full load I_,I r ,I L = per phase stator, rotor and load
power output. The study would involve current
modelling, computer simulation, analysis, V L , V M , Vg = load, machine terminal and airgap
identification of constraints and estimation voltage
of desired parameters. Further, comparison of
SEIG performance under short and long shunt (all reactances referred above relate to the
configuration to assess their relative merits base frequency)
and demerits is also desired.
In this paper a simple and general The magnetization characteristic of the
mathematical formulation for short shunt machine is of prime importance in the analysis
configuration has been presented. From the of SEIG. Figure 2(b) shows the characteristic
equivalent circuit parameters a method to of the machine considered, relating the ratio
evaluate the performance characteristics has of airgap voltage to frequency (Vg/F) with
been explained. A simple methodology has
been proposed for the selection of suitable
values of capacitors to obtain minimum Vg/F (p.u.)

regulation. The effect of capacitors on

performance quantities like winding current
and machine voltage has been discussed so as
to evolve guidelines for designing the
system. A series of tests have been performed
and relevant experimental results are
presented along with theoretical ones. It is
shown that once the mathematical foundation
is laid and the methodology is understood the
design of a self-regulating SEIG becomes Xm (p.u.)
relatively simple. The study is extended to Fig. 2(b) Variation of V o /F with X*

long shunt configuration and the two

configurations are compared. the magnetizing reactance Xm. The saturation
portion of t h i s characteristic can be
linearised and expressed mathematically in the
THEORY form [7] :

The present study uses the standard V_/F (1)

steady state equivalent circuit of the SEIG
with the usual assumptions [7,12,16], Where Kj_ and K2 are constants dependent
considering the variation of magnetizing on the design and material of the machine.
reactance with saturation as the basis for For the purpose of obtaining the required
calculation. The equivalent circuit of SEIG operating point for the given values of X c h
with capacitors connected in short shunt and X c e , the only unknown parameters for a
configuration is shown in Fig. 2(a). given speed and load are Xm and F. To
evaluate these parameters the loop equation
where for the current (Is) can be written as
R
s' R r = per phase stator and rotor resistance
= per phase stator and rotor leakage T * = n (2)
reactance
x
m = magnetizing reactance where
x
ch = per phase capacitive reactance of the
shunt capacitor, C sh (3)
12

and SELECTION PHILOSOPHY OF ELEMENTS

When the SEIG is operating under no load

Z2=RrF/(F->))+jFXr,
Since under steady state operation of
SEIG, Is cannot be zero,
This equation after separation into real
and imaginary parts, can be rearranged into
the following two nonlinear equations which
are functions of Xj, and F; denoted as J(Xffl,F)
and $(X m ,F).
f(X m ,F)=(C 1 X m + C 2 )F 4 + (C 3 X m + C 4 )F 3
+(C 9 X m +C 1 0 ) = 0
condition, there will not be any current
(4) through the series capacitor (C se ) and only
shunt capacitor (C sn ) will be effective in the
circuit. Therefore, effect of C s h is reflected
on the no load performance of the SEIG. But
when loaded, both shunt and series capacitors
(5) will be effective. Hence, proper value of
these elements can be chosen by first studying
the variation of no load terminal voltage with
C s n . Having chosen a suitable value for C s n
from this curve, the influence of C s e can be
studied by observing the effect of C s e on
voltage regulation of the SEIG. Appropriate
value of C s e can be selected from the range of
values thus obtained depending upon the
desired regulation and other operating
constraints. Following section illustrates the
procedure.
(7)

RESULTS AND DISCUSSION

t( X m' F )=(D 1 X m +D 2 )F 3 +(D 3 X m +D 4 )F 2
A standard three phase induction machine
+(D5Xm+D6)F+(D7Xm+D8) = 0 (8) with the specifications detailed in Appendix-
II has been considered for the present study.
Where C1,c2,...,Clo and DlfD2,
constants defined in Appendix-I. Selection of Shunt Capacitance:

The variation of no load terminal voltage

V m o with shunt capacitance (C sn ) is shown in
These two equations can be solved using Fig. 3(a). It is seen that V m o increases with
any of the suitable numerical techniques to C s n . Appropriate value of shunt capacitance
obtain the values of Xm and F. In the present can be chosen depending on the maximum
paper Newton - Raphson method [7] has been permissible voltage across the machine
employed to solve these equations. terminals and the voltage requirements on
part loads. As a first approximation, value of
C s n (16.7 /*.F) corresponding to no load
After obtaining Xm and F by choosing terminal voltage equal to rated value (1.0
suitable initial values, Vg can be computed p.u.) has been considered.
from equation (1). Then the following
relationships can be used to compute the Selection of ries Capacitance:
generator performance at a particular load and
speed: Figure 3(b) depicts the variation of full
load voltage regulation with series
= V[Z1 (9)

No load voltage, V m0 (p.u.)

Ir = Vg/[RrF/(F-V) + jFXr] (10)

(11)

(12)

L
L^L (13)

Input power P i n = -3I r 2 R r F/ [F-S> ] (14)

7 e 11 13 W 17 19 21 23 25 27 29 31 33 35
Shunt capacitance C s h </iF)
Output power P o u t =3I L 2 R L (15) Fig. 3(a) Effect of shunt capacitance on no load voltage
 13
% Regulation

Cse(/<.F) Output Power (p.u.)

Fig. 3(b) Effect of series capacitor on voltage regulation Fig.4(a) Variation of output voltage with load

capacitance obtained through both

experimentation and computation. Here the
V* (p.u.)
voltage regulation is defined as the
percentage change in the load voltage as the
generator delivers power from zero to rated •

value. Usually variation of terminal voltage

with load is permitted within a specified _
band. Taking the permissible voltage
Computational Experimental Cse
variation of ±6%, the range of C s e for this
1 0 20/<F
band is shown in the figure. Hence, a range 2 40/(F
of C s e ( 3 0 - 7 0 MF) is available for the 3

selection. It reveals that there is a

1 1.2
distinct minimum value of voltage regulation, Output Power (p.u.)
Fig.4(b) Variation of machine winding voltage with load
the corresponding value of C s e is the best
value as far as full load voltage regulation
is concerned. Further, a close agreement
a (p.u.)
between computed and experimented results
can be noted. 1.2
©\ ®
1
Effect of C s e on Vm and I_s:
0.S o °^<^S
0 ^
It is seen that once the mathematical 0.6
equations are formulated, selection of proper 0.4
Computational Experimental Cse

values of capacitive elements of short shunt

SEIG for minimum regulation becomes very 0.2 © 40 ^ F
*
simple based on the procedure already 0

explained. Having identified this method for Output Power (p.u.)

Fig.4(c) Variation of winding current with load
the selection of capacitors, it is important
to see how best this selection is, as far as
the other relevant performance indices of the the cost of high capacitance and poor
machine are concerned. regulation.
An important criterion is that the However, with the value of C s e (45 j^F)
machine should deliver needed power at desired chosen from the proposed analysis, performance
voltage levels without exceeding the of the machine is not so much constrained.
permissible electrical and magnetic loading. Figure 5 shows various characteristics of the
The effect of C__ on load voltage (VT) short shunt SEIG for the chosen values of the
x
Sc -* J_i'
capacitor elements. It is seen that at rated
and machine terminal voltage (VM) is shown in
current, the output of the generator is more
Fig. 4(a) and 4(b). It is observed that there
than the rated value. V c e is the voltage
is a rise in the value of machine terminal
across series capacitor which increases almost
voltage for lower values of C s e , where as,
linearly with load. For the selected pair of
higher values of C g e results in poor c
sh a n d c se L t i s s e e n t h a t the full load
regulation.
voltage regulation is as low as 2%. For these
Figure 4(c) shows the variation of values of capacitors, the generator can be
winding currents with output power. It is seen loaded up to 107% of the rated power with the
that for lower value of the power
capability of the SEIG reduces since at rated winding current less than the rated value.
winding current it can deliver less than This overload capability enhances the
rated power, needing the machine to be robustness of the SEIG. The voltage across the
derated. Where as in the case of higher C s e machine terminals has been found to be only
the power capability of the SEIG increases at 8% above the rated value at rated power
14

output. This over voltage is not considered as Short versus Long Shunt Connection:
a matter of serious concern, as it is within
acceptable range for normal machines. Machine Long shunt configuration is yet another
can however be designed to withstand such a way of improving voltage regulation of SEIG,
over voltage. The low value of voltage wherein series capacitors are located between
regulation makes the SEIG attractive when the machine terminals and the shunt
compared with synchronous generator, which capacitors. The theory described for short
is not only expensive but also has complex shunt SEIG is similarly applicable for this
control circuits. configuration also.
Experiments carried out on short shunt
SEIG were repeated for long shunt SEIG. A set
,Vc e and Is (p.u.) of predicted and experimental results for
V
M _-- long shunt SEIG is shown in Figure 7.
Approximate values of capacitors for this
particular configuration can be obtained from
c
»e- 46.0 /»F the graphical method as suggested by Bim etal
16.7/IF
— [14].
It is seen that like short shunt SEIG
'
.
— long shunt SEIG is also self regulating in
•

nature. A low voltage regulation of 6% is

— •

obtained for the values of the capacitor

Output power (p.u.)
Fig.S Characteristics of short shunt SEIG
Since Csh is also known to have some
effect on voltage at full load, considerable
complexity is involved when the choice of C s n
is not restricted merely to no load studies.
Appropriate criterion for the analysis needs
to be formulated to see if the choice of C s n
resulting in V m o above or below 1.0 p.u. can
give acceptable and improved characteristics.
Further, similar to the range of values of
C s e , a range of values of C s n is also
available for which the no load and part load
voltage lies within the permissible voltage
band. But as seen from Fig. 6, for lower
values of -sh dips are noticed in the
elements. To achieve this voltage regulation
series capacitor of 100 f*-F is required. This
value of series capacitor is more than twice
that required in short shunt configuration.
This involves additional expenditure. Further,
the series capacitors need to be rated for
higher current, as they carry machine line
current, where as in short shunt configuration
they carry only load current. It is also seen
that the machine terminal voltage remains at
considerably higher values over the whole
operating range.
It is thus concluded that short shunt
configuration is superior compared to
long shunt, both in terms of economy and
performance of the SEIG system.

VL (p.u
1.2

Cse • AB.OjtF 1<

~—• * * V
L » »

0.8
Jsh • 18.0/tF
I.
0.6
Cse • 100>F
0.4 Csh • 18>iF

0.2 0 • • Experimental
Computational

0 0.2 0.4 0.6 0.8 1 t

Output Power (p.u.)
Fig.7 Characteristics of long shunt SEIG
Output Power (p.u.)
Fig.6 Effect of Shunt Capacitance on load characteristics

generated voltage on partial load conditions.
Therefore, while selecting the values of C s n
and C g e apart from the full load regulation,
machine terminal voltage and the winding
currents, voltage dip at partial loads becomes
a constraint in the system design.
CONCLUSIONS
The short shunt SEIG has been studied in
detail and a suitable methodology has been
developed for the selection of appropriate
values of capacitor elements. Using a simple
mathematical technique its various operational
 15
features have been discussed highlighting the REFERENCES
constraints imposed on the system design. It 1. E.D. Bassett and F.M. Potter,"Capacitive
is seen that voltage regulation of as low as Excitation for Induction Generators",
2% is possible to achieve by choosing suitable AIEE Trans. (Electrical Engineering), Vol
values of capacitive elements using the 54, pp. 540-545, 1935.
proposed methodology. The agreement reached 2. C.F. Wagner, "Process of Self-excitation
between the theoretical and test results, of Induction Motors", AIEE Trans. (Elect-
confirms the validity of the analysis carried rical Engineering), vol.58, pp.47-51,
out and the criteria adopted. The technique is February 1939.
extended to study long shunt SEIG and a 3. B. C. Doxey, " Theory and Application of
comparison of the two configurations has Capacitor Excited Induction Generator",
established the superiority of short shunt The Engineer, no.29, pp. 893-897,
configuration. November 1963.
It is expected that this study will 4. D.W. Novotny, D. J. Gritter and G.H.
further open up new avenues resulting in Studtmann, " Self Excitation in Inverter
detailed and useful investigations for Driven Induction Machine", IEEE Trans, on
extending the applications of self regulating Power Apparatus and Systems. vol. PAS-96,
SEIG. no.4, pp 1117-1125, July August, 1977.
5. M. B. Brennen and A. Abbondati, "Static
Exciter for Induction Generators", IEEE
APPENDIX-I Trans, on Industry Applications, vol. IA-
13, no.5, pp. 133-139, September, 1977.
Assuming X s =X r =X^ 6. J. Arrillaga and D.B. Watson, " Static
Power Conversion From Self Excited Induc-
C 1 =-2.X 1 .R L , C 2 =-X 1 .RL, C 3 =-C 1 , C 4 =7C 2 . tion Generators", Proc. IEE. vol. 125,
no. 8, pp. 743-746, August, 1978.
c
5= x ch- R L-( x ce- x ch)•( R s +R r>' 7. S. S. Murthy, O.P. Malik and A. K.
Tandon* "Analysis of Self Excited
C
6=( x l- X ch +R s- R r>- R L-( x ce- x ch)•( R s +R r>- x l' Induction Generators", Proc. IEE. Part C,
Vol. 129, pp. 260-265, November,1982.
C 7 =-V.X c h .R L + V.(X c e -X c h ).R s , 8. J. M. Elder, J. T. Boys and J. L.
Woodward,"Process of Self Excitation
C 8 =X 1 .C 7 , C 9 =0.0, C 1 0 =X c e .X c h .R r Induction Generator", IEE Proc. , Pt. B,
vol. 130/2, pp. 103-108, March 1983.
D
l=- 2 - (xce-xch> -Xi+CRs+V -RL< 9. O.P. Malik, D. Diwan, S.S. Murthy, T.
Grant and P. Walsh, " A Practical Solid
State Voltage Regulator for Self Excited
D 2 =(R S +R r ) . X - L . R L - C X ^ - X
Induction Generator", Proceedings of IEEE
Industrial and Commercial Power system
D3=2.V . (X c e -X c h ) .X-L-y.R Conference. Milwaukee (USA), June 1983.
10. G. Raina and O.P. Malik, "Wind Energy
D4=y- (X c e -X c h ) . X ^ - V . X ^ ,, D 5 =X c h .X c e ,
Conversion Using a Self Excited Induction
Generator", IEEE Tans, on Power Apparatus
D
6=(xl-Xce-Rr-RL>^ch+^ h)- R r - R s' and Systems. vol. PAS-102, no.12,
pp.3933-3936, December, 1983.
D
7=-V- X ce- X ch' D
8=DVX1 11. J.M. Elder, J.T. Boys and J.L. Woodward,
"Self Excited Induction Machine as a
Small Low-Cost Generator", IEE Proc..
APPENDIX-II
vol. 131, Pt. C, No.2, pp. 33-40, March
1984.
Base Quantities
12. S. S. Murthy, H.S. Nagaraj and Annie
Kuriyan, "Design Based Computational
Power - 3.7 kW, Voltage/phase - 415 Volt
Procedure for Performance Prediction and
Current/phase- 4.39 Amp, Frequency- 50 Hz
Analysis of Self Excited Induction
Speed - 14 3 0 rev/min.
Generators Using Motor Design Package",
Proc. IEE. Vol. 135, Pt. B, No.l, pp. 8-
Equivalent Parameters 16, January,1988.
13. S. S. Murthy, B. P. Singh, C. Nagamani,
Rs = 0.053 p.u., Rr = 0.061 p.u.
K.V.V. Satyanarayana, "Studies on the Use
Xs = Xr = 0.087 p. u.
of Conventional Induction Motor as Self
1^ = 1.6275, K2 = 0.3418
Excited induction Generator", IEEE Trans.
16
on Energy Conversion, Vol. 3, No. 4, Dec.
1988, pp 842-848.
14. E. Bim, J. Szajner and Y. Burian,"Voltage
Compensation of an Induction Generator
with Long Shunt Connection", IEEE Trans,
on Energy Conversion. Vol. EC-4, pp. 526-
530, Sept. 1989.
15. Bhim Singh, S.P. Singh and M.P. Jain,
"Design Optimization of Capacitor Self
Excited Cage Induction Generator",
Journal of Electrical Power System
Research. Vol. 22, No.l, pp. 71-76, 1991.
16. S.P.Singh, Bhim Singh and M.P. Jain,
"Performance Characteristics and Optimum
Utilization of Cage Machine as Self
Excited Induction Generator", Paper no.
90 SM 284-0 EC, Presented at IEEE PES
Summer Meeting, Minnesota, 1990.
17. L. Shridhar, Bhim Singh and C.S. Jha,
"Improvement in the Characteristics of
Self Excited Induction Generator", Paper
no. 92 WM 02 0-8 EC, Presented at IEEE PES
Winter Meeting. New York, 1992.
18. Bhim Singh, R.K. Mishra and M.K.
Vasantha, "Voltage Regulator for .Isolated
Self Excited Cage Generators", to
appear in Journal of Electric Power
System Research, vol 24, no.2, pp. 75-
83, 1992.
BIOGRAPHY
L. Shridhar was born in
Bhilainagar, MP (India) in
1966. He received his B.E.
H * $ degree from Maulana Azad
1
College of Technology,
Bhopal, and the M.Tech
* * degree from Institute of
Technology- Banaras Hindu
University, Varanasi.
He joined the department of
electrical engineering, Indian Institute of
Technology, Delhi in July 1990 and is
presently, a full time research scholar with
the department working, towards his Ph.D.
degree. His areas of interest are computer
based analysis and design of electrical
machines and their efficient energy conversion
and application in non-conventional power
plants.
Dr. Bhim Singh, was born at
Rahamapura in U.P. in 194 6.
He received his B.E. degree
from Roorkee University,
and M.Tech and Ph.D. degree
from IIT-Delhi in 1977,
1979 and 1983 respectively.
From 1983 to 1990 he was with the
department of electrical engineering,
University of Roorkee. At present he is
Assistant Professor at Indian Institute of
Technology, Delhi.
He has over 60 papers to his credits in
the field of CAD, Power Electronics and
Analysis and Control of Electrical Machines.
Prof. C. S Jha, was born
at Vijainagar in Bihar
(India) in 1934 and
educated at Patna Universi-
ty, Indian Institute of
Sciences, Bangalore, Heriot
Watt College, Edinberg
(U.K.), and Bristol Univer-
S ity (U.K.). Has been a
rroressor of Electrical Engineering at Indian
Institute of Technology- Delhi since 1964.
He has made significant contribution in
electrical machine theory and applications and
published a large number of papers. Has been
involved in planning and administration of
technical education in India since early
1970s. He was Director of the prestigious IIT
at Kharagpur (1974-78), was Educational
Advisor to the Government of India on
technical education planning and has been
active in curriculum planning and development
of engineering education in India. Has been
Visiting professor in many universities in the
West, a member of the board of Trustees of
Asia Institute of Technology, Bangkok (1974-
86) and is a member of UNESCO international
meeting group on continuing education of
engineers since 1975.
At present, he is the Vice Chancellor of
the Benaras Hindu University at Varanasi.
Prof. B.P. Singh was born
in Singhiya, in 1940. He
received his B.Sc. (Engg.)
degree in 1963 from BITS,
Sindri, ME in Electrical
Engg. in 1966 from Calcutta
University and Phd. in 1974
from IIT Delhi.
He was a Senior Fellow at
BE College, Howrah (1963-
1966) and after serving MIT Muzaffarpur as a
faculty member over a decade (1966-78), he
joined IIT Delhi in 1978, where he is a
Professor with the Dept. of Electrical Engg.
He was a visiting Professor at California
State University, Long Beach during 1988-1990.
His research interests are in design, analysis
and control of electrical machines.
Prof. S.S.
born in Karnataka in 194 6.
He received his B.E. degree
from Bangalore University,
M.Tech from IIT Bombay and
Ph.D. degree from IIT-
Delhi in 1967, 1969 and
•* J»
1974, respectively,
v • -*. ^ i- He was at BITS Pilani
during 1969-70 and has been at IIT Delhi
since 1970 where he is a Professor with the
Department of Electrical engineering, he was a
visiting staff at the University of new Castle
Upon Tyne during 1975-76 and at the
University of Calgary during 1980-82. Has held
the post of Director at ERDA, Baroda (India)
during 1990-92.
He has published a large number of papers
and has edited two laboratory manuals. He has
executed many industrially oriented
consultancy projects. His areas of interest
include electrical machines and drives,
efficient electrical utilization and
engineering education.

Discussion
0. P. Malik (The University of Calgary, Calgary, Alberta,
Canada):
The authors have presented a very interesting paper which
is a good step forward towards improving the feasibility of the use
of self-excited induction generator for electric power generation.
This discussor would like to ask the authors the following two
questions:
1. In the analysis and all the studies described in this paper,
the authors take the load to be resistive. Purely resistive
load is not the most common in practice. Have any studies
been done with reactive loads? How does the load power
factor affect the performance characteristics such as voltage
regulation, and the value of series and/or shunt capacitance
required?
2. Use of series capacitance opens up the possibility of
resonance and all its attendant effects some of which can
cause heavy damage. Have any studies been done to
investigate this effect either for steady-state or transient
conditions?
Satisfactory answers to questions like the above would
certainly enhance the applicability of self-excited induction
generator.
Manuscript received February 16, 1994.
17
L. Shridhar, B. Singh, C. S. Jha, B. P. Singh, and S. S.
Murthy:
The authors would like to thank Prof. O.P.
Malik (The University of Calgary) for his interest
in our paper. The points raised by him are
extremely important. Since submitting the paper
we have extended the study to reactive loads. It
is found that the requirement of series
capacitance increases with lower power factor
loads to keep the voltage regulation within
permissible limits of +_ 6%, Vs ^ 1.06 and Is < 1.0
(for 0.9 p.f. lagging load, the minimum value of
series capacitor became 75 ,uF as against 60 uF for
u.p.f load). The shunt capacitor is not affected
much and the same value may be used for reactive
loads as for resistive loads. The voltage
regulation feature of the series capacitance with
inductive loads is thus seen to be as effective as
with resistive loads. The transient behaviour of
the short-shunt configuration for static loads
have been found quite satisfactory. However,
while running a motor loadr we did come across
occasional resonance under short-shunt
configuration due to which the motor drew large
currents and ran at much lower than rated speed.
This phenomenon is being investigated and it is
hoped to present suitable remedy for this in a
future publication.
Manuscript received August 3, 1994.