Precise Method f o r Measuring Xd and Xq Based on Slip Test

o f Synchronous Machines

Osamu. SUGIURA Yu j i. AKIYAMA

Dept.of Elect.and Comp.Science. Yember IEEE
University of Yamanashi Dept. of Elect. Engrg.
Yamanashi, Japan Kanagawa Institute of Technology
Kanagawa, Japan

ABSTRACT R, :resistance of stator winding.

An experimental method for measuring the direct L :self inductance of stator winding.
and quadrature axis synchronous reactances (Xd and M :mutual inductance between the stator windings.
xq) of polyphase synchronous machines with salient XI :mean value in self reactance of stator winding.
pole is newly developed. It is based on a measuring Xo :maximum value of variation in self reactance
method of slip test. It can be applied to all the of stator winding.
synchronous machines, but primarily it is developed Xd :direct axis synchronous reactance.
and proved for reluctance motors. Precise measuring xq :quadrature axis synchronous reactance.
method was derived by the use of tensor analysis 0 :angle between stator winding of a-phase and d-
method, and was introduced to minimize measurement axis of rotor.
errors due to the stator resistance (Rs) as well as a :angle between stator winding of a-phase and d-
the machine losses. Detailed results are presented axis of rotor at time t=o.
in case of one reluctance motor which has a highly o :supply frequency, rad/s.
variable Rs. In fact, correlation with other o r :rotating speed, rad/s.
methods is proved to be quite good, The effects of p :differential operator.
Rs are observed in the tested motor. t i : instantaneous torque.
t m :mean torque.
7 :pulsating torque.

I . INTRODUCTION Subscripts
s :stator.
The various methods to measure Xd and xa of the d :direct axis.
synchronous machines have been used in many ways in q :quadrature axis.
the past, as in [l] - [6] . The slip test has
widely been used to measure Xd and xq. By this
a, b, c :symbols of 3-phase.

simple experimental test,the proper measured values II. ANALYSIS OF RELUCTANCE YOTOR
of Xd and x4 can be obtained. But, this measuring
method takes no account of R3 when the values of Xd Applications of the theory of tensor to analyze
and xq are computed, However, Rs can not be ignored rotating machines were developed by G.Kron, as in
in small motors because the resistance is larger in [71 . This Kron’s method is expedient to analyze
small motors than in medium and large motors. two phase machines, because the two hypothetical
In this paper, the authors made a try to take R3 windings are assumed to be two phase windings of
into consideration. The measuring method of Xd and the machine under consideration. But, for the three
xq was to make the theoretical analysis at synchro- phase machine, it is necessary to transform the
nous speed of the reluctance motor. This method is three phase actual windings into the hypothetical
obtained by applying the analytical results, and is windings of the primitive machine with the aid of
experimentally proved. connection matrix, which will make the resultant
impedance matrix rather complicated.
List of principal symbols To avoid such complicated manipulation, Poly Axis
Matrix Yethod developed by T.Takeuchi as in [8] is
1462-x/93$03.OO Q lW3IEEE
0-7803-
155
 h L

-?----
Fig. 3. Variation of Yab with 6 .

I cos20 1

Fig. 1. Winding model of RI. And, the mutual inductance ?dab between the stator
windings changes sinusoidally as in Fig. 3. Then,
the mutual inductance between the stator windings
is written as

0
-
Fig. 2.
*/2
L
I
7~ -e
Variation of La with 8 .
The voltage equations of stator windings are shown
as a matrix formula as in (3).

:[ ] Rs+PLo $Mob, Pn&c

= [DMbaj
t

R s + P L , .Mbc][:!
I j (3)
used in this paper. This method analyzes matrix uC PfiIcm PMcb, Rs+PLc
equations on the actual axes of windings and fa-
cilitates the analysis without any confusion. Transforming (3) by the three phase sequence trans-
formation matrix A 3 , we have the voltage equation
A. Voltage Equation of (4) based on the symmetrical co-ordinates axis
as
Fig. 1 shows the winding model of the reluctance
motor(R.'d) in which ai, bl and CI are three windings
spaced by 120" each other on the stator. La, L b and
Lc are the self inductance of the stator windings,
and Mat,, l b c and Mca are the mutual inductance
between the stator windings. The salient pole rotor
is rotating with an angular speed of w in the
clockwise direction as shown by an arrow wt in the
Fig. 1.
The self inductance La of the stator winding
changes sinusoidally as in Fig. 2. Then, the self
inductance of the stator windings may be written as

156
were B. Stator Current
The impedance matrix of (7) is the stationary
impedance matrix. Thus, we can get the inversion
of (7) as

where
The zero sequence elements are putted aside from
(41, because the grounding fault do not consider.
BII-.Rs+ ( L + M ) ( p - j w ) - (9) (p-tjo)

Then, the resultant voltage equation of (5) is

written as

The instantaneous three phase voltages are written

as
The impedance matrix of (5) is a function of time
t. To eliminate 8 from the elements of this
impedance matrix, we define the commutation matrix
K as
Then, the voltages based on the symmetrical co-
ordinates will be written as

which can be chosen by inspecting (51, as in [8] .

Transforming (5) by the commutation matrix K
minding that

Substituting (9) in (8) obtains (10).

where

The impedance matrix of (7) is the stationary

impedance matrix which is no more function of time
8=wt+a. And, the currents based on the symmetrical co-
ordinates will be written as

157
 Fig. 5. Vector diagram of I a .
The relation between a and the power angle 8 can
be written as
And, the instantaneous power Pi will be obtained
by premultiplying [i]t to (16).

Substituting (13) in a of (12) obtains (14).

When we can transform the above equation into the
symmetrical co-ordinates by the sequence matrix A,
Pi is written as
And, the self reactance of the stator winding Xa
versus the angle 8 is showed by Fig. 4.
From (14) for the instantaneous stator current,
the stator current in vector form 1, is written as

in which Ath is expressed as

Therefore, Fig. 5 shows the vector diagram of Is

at arbitrary power angle a.

C. Instantaneous Torque CZ’(,)l is the symmetrical matrix, and is expressed

as
The voltage equation of any machine is given by C~’~P)I=CR’]+PCL’]
the matrix as
CUI = CZ(,,lCiI (16) where ER’] is the resistance matrix, and CL’I is the
inductance matrix.
For the sake of simplicity, we may write

1 * [ R ’ ]= [I?,], 1 *CL‘]=EL,]

and note that these both matrices are symmetrical.

Then, (17) can be written as
p , = [i ’ ] t - l * [ Z ’ ( p ) ] [ i ’ ]
[ [ <’It 1.p[ L’ ][ i‘]
= [ i ’ I I - 1 . [ R ’ ]i’]+
0 v 2 -e
= [ i ‘ l f~ 1 ~ C R ’ I C i ’ I + C ~ (’ [l L
~ ’- ]1p~[i’]+(p[L’])[i ’ ] )

Fig. 4. Variation of Xa with 8 . = [ ~ ’ I ~ C ~ ~ ~ ~ ~ ~ I + C ~dCi’3 ~ - )C; Ti’]; -] + (18)

’ I ~ [ ~(~drL,1
~it

158
 (20) rr =-.
3 xovz sin2(3 =-.3 (zt%-dV2.sin26 (25)
w x+-xo2 w 2xdxq
In the right hand side of the above (201, the
first term is the power expended in heating the Equation(25) is the general torque expression of
conductors. And the second term is the rate at RM.
which energy is being stored in the magnetic field,
that do not contribute mechanical output. Also note IU. DERIVATION OF METHOD FOR MEASURING xd AND Xg
that in the above (20) 1
K=$i’] c[L.][i’] Fig. 6 shows the vector diagram of I, at the angle
of phase difference dm. In Fig. 6, I m i n is the
is the instantaneous stored energy in the magnetic vector of the minimum stator current with the angle
field. of phase difference dm.
Therefore, the third term must represent the Fig. 7 shows the variation of W, I and V with the
mechanical power expended by the rotor. And, the power angle 6.
power may be expressed as The waveforms of input power, stator current and
impressed voltage are the envelopes of the instan-
taneous maximum value. If RN is rotating nearly at
synchronous speed, it is permitted to exchange time
t for 6. Accordingly, the variation of I and V is
On the other hand, z i is an instantaneous torque observed quite the same as experimental data ob-
and o is an angular speed of the rotor. Then, tained by the slip test.
o z i will be the mechanical power expended by the Then, the impedance 2, i n of RM which is viewed
rotor, thus, we have from power source is

(21)

This is the general expression of torque obtained I

from output. where

L+M
~(L0-2Mo)c’Z~
2 3
(23) Fig. 6. Vector diagram of I,.

159
 Substituting (30) in (31) obtains (32)
W 4 - Wm;n
- - 4I h W
Xo=JRs2+Xi2
I
I
max-

max+
I
I
m i n

min
] (32)
v
I Then, the direct and quadrature axis synchronous
reactances Xd and xq will be

I' X d = X l + X O , xq=xI-xO (33)

Hence, Xd and xq is derived from three kinds of

measurement as follows.
(a) The resistance Rs is measured by a simple d,c
Fig. 7. Variation of W, I and V with 6. test.
(b) The machine under test is driven steadily at
synchronous speed, and its impedance Z m i n is
measured for the rotor angle which gives the mini-
mum current at the terminal (see Fig. 6).
were
Wm;n Wm;n
Rmin =
3 I m i n e 1

The tangent of (26) is

Thus, this test gives value of XI as follows.
t a n (6m) = XIV/A -- - =X -t X m i n
(27)
R s V / A Rs Rmin

Thus, comes the following equation as (c) Being used the slip test, I m a x and I m i n which
are a maximum and a minimum currents, are measured
to get the value of Xo.
Fig. 8 shows the oscilloscope waveforms of V and
From Fig. 6, two currents of I m i n and Ima, can I.
be written as From the measured data,

X D = J R , ~ + X I ~ [I max-

I m a x f
I
I
min

min 1
Then, two currents of 111 and 112 may be defined
as terminal voltage
I pax+ I m i n I I
I l l =
2
(30)
I tax- I min
I12=
2
From (14), we have
f stator current I

-I l -
l - J Rs2+Xi2 (V/A>
r 12 xo (V/A>
-
- 4 Rs2++X1* (31)
x o Fig. 8. Oscilloscope waveforms of V and I.

160
 The authors investigated the effect of Rs on the
computed values of xd and Xg by using the equiva-
lent circuit in Fig. 10.
It is assumed that xd5 is the direct synchronous
reactance which is computed by the slip test. And,
Xd is the true direct synchronous reactance of RN.

W
w2 'KO - Then, X d s will be

(34)
Fig. 9. Circuit for measuring Xd and xq.
Changing the above equation into a more suitable
form.
Then, the value of xd and xq are obtained as
Xd=xl+xO
xq = X I - xo Where, K is the error factor, and can be written
as
Fig. 9 shows the experimental circuit for measur-
ing XI and XC, thus, Xd and xq.
K=d 14- (Rs/Xd) ' (36)
In Fig. 9, Rg was inserted in the armature circuit
for the experiment to determine whether the precise Fig. 11 shows the variation of K with (Rs/Xd).
slip test for measuring Xd and xq is good or not. Differentiating K with respect to (R,/Xd) to in-
vestigate the characteristics of the curve in Fig.
IV. EXPERINIENTAL RESULTS 11.
d K
The effect of the stator winding resistance R, on K' =
the computed values of xd and xq is investigated in d (Rs/Xd)
the first place. Then, the authors computed ap-
proximately how large the calculative error is due
to Rs.
In the second place,the experimental results based K" = 1 / ( 1 f ( R s / x ~ )2 } 3'2> 0
on the precise slip test are taken into consider-
ation, and a comparison of the conventional slip K increases always with (Rs/Xd). When K is dif-
test and the precise slip test with regard to the ferentiated twice with respect to (Rs/Xd), it can
effect of R, has been made. be obtained from K" that the curve of Fig. 11 in-
creases rapidly with (R~/xd).
A. Calculative Error in the Slip Test
6
Fig. 10 shows the approximate d- axis equivalent
circuit of RI at power angle 6=0.
4
K

Fig. 10. Approximate d-axis equivalent circuit

of RM. Fig. 11. Variation of K with (Rs/xd).

161
B. Precise Slip Test 1 A -Precise
B -slip test
slip test

The authors carried out the experiment according 20

Xd
to the results of the precise slip test which was
described in Chap. NI. Further, the authors carried
out the conventional slip test which is based on
the following assumption; when the rotor is in the
direct and also in the quadrature axis position,the
currents drawn by the machine are a minimum and a
maximum respectively. t
Fig. 9 shows the circuit for measuring Xd and xq.
The test motor is rated 1.0kw, 200 volts, 50 Hz.
The additional resistance Rq was connected with the
circuit of stator winding to investigate the effect
of the stator winding resistance R3, and the experi- Fig. 12. Comparison of the calculated values of
ments were carried out by changing the values of R q , Xd and Xq with the test results
Fig. 12 shows these results. Test results of Xd obtained by two different tests.
and Xq obtained by these methods for the test ma-
chine are plotted as function 01 ( R 3 t R s ) in Fig. 12.
Fig. 12 shows a comparison of the calculated values his helpful advice and the ldachine Shop Group of
of Xd and xq with the test results obtained by two Yamanashi University for producing experimental
different tests. An increase in REI results in an devices.
equivalent increase in Xd and xq in the slip test,
and the values are almost always constant in the
precise slip test.
REFERENCES
Therefore, the precise slip test confirms the va-
lidity of a measuring method for Xd and xq. P.J. Lawrenson, and 1. A. Agu, " Theory and performance of
poly-phace reluctance machines, " Prm.IEE, vol. 3, pp. 1435-
V . CONCLUSIONS 1445, Aug 1964.
V.B. Honsinger, " The inductance L d and Lg of reluctance
machines," IEEE, vol. PAS-90, pp. 298-304, Jan/Feb 1971.
A new measuring method of x d and Xq which took R. W. Henzies, R. 1. Nathur. and H. 1. Lee, a Theory and oper-
account of the stator winding resistance was pro- ation of reluctance motors with magnetically anistropic
posed by the authors. The authors carried out the rotors, U -synchronous performance," IEEE, vol. PAS-91, pp.
experiments according to this measuring method and 42-45, Jan/Feb 1972.
E. A. Klingshirn, " DC standstill torque used to measure Lg
examined the results. The method of calculating of reluctance and synchronous machines," IEEE, vol. PAS-97,
the reactances gives results which agree well with pp. 1862-1869, Sep/Oct 1978.
measured values of unsaturated state. A.A. Fock. and P.M.Hart, " New method for measuring xd and
The method described in this paper for the Xg based on the P-Q diagram of the lossy salient-pole
machine," PIEE, Pt.B. vol. 131. pp.259-262. 1984
measurement of salient-pole machine reactances has T. Fukao, " Principles and output characteristics of super
more advantages than the slip test, and the ex- high-speed reluctance generator system," IEEE. vol. IA-22.
periments of this method are widely acceptable and pp. 702-707, Jul/Aug 1986.
is easier in comparison with other methods. G. Kron, " The application of tensors to the analysis of
rotating electrical machinery," Gen. Elec. Rev. Aug 1935.-
May 1938.
ACKNOWLEDGMENT T. Takeuchi, " Matrix theory of electrical machinery," The
OHM-SHA LTD., 1967.
The authors would like to thank Dr. H. Kazuno for

162