Metho: Abstract
Metho: Abstract
Metho: Abstract
C.-M. You ng
C.-C. Liu
C.-H. Liu
31
Q1' off switches are reversed. The same modulation method
-1
is applied to Sds(t)and Tr(t),in which the correspond-
main ing switching elements are Q2 and Q3 in Fig. 1 and (2;
t to Qlin Fig. 2. Fig. 3 shows the waveforms of S,,(t),
-vc
I- Sdr(t)and Tr(t).The inverter output waveforms, V,,(t)
and Vds(t),are also shown in Figs. 4 and 5. It should
be noted that both inverters have the same output
characteristics, only the input rectifying characteristics
can be used to differentiate the two inverters. To eval-
4 uate the performance of the two inverters on the input
I 1 I I I I I
- 2 Inverter circuit for comparing with Fk. I rectifier side, the performance indices such as the
source total harmonic distortion (THD), the power fac-
tor (PF) and the ripple factor (RF) are employed [12].
The ripple factor is used to evaluate the voltage wave-
form across the capacitors. Table 1 summarises the
simulated results when the inverters are operated at 60,
45, and 30Hz, respectively. Fig. 6 and Fig. 7 show the
respective source current waveforms, is(@ The capaci-
tor voltage waveform, Vcl(t),in Fig. 1 is shown in Fig.
0.80 0.81 0.82 0.83 8, while the capacitor voltage waveform, Vc(t),in Fig.
-3 SP WM technique 2 is shown in Fig. 9. Comparing the results given in
Waveforms of Sqs,S,,, and T, Table 1 and Figs. 6, 7, 8 and 9, the inverter shown in
Fig. 2 clearly performs better than the proposed
200 1 inverter in this paper. The reason is clear, since the
proposed inverter has fewer power elements than the
inverter of Fig. 2. However, the performance of the
proposed inverter can be improved by incorporating
the power factor correction technique.
two input rectifying diodes in Fig. 1 by two power
-200 I I I I I I I
transistors and employing an appropriate power factor
0.80 0.81 0.82 0 83
correction (PFC) circuit, such as the one given in [13],
SP WM technique the THD, PF and R F can all be improved significantly,
Inverter output V, which are also tabulated in Table 1.
-10 j I I I I 1 I I I
2-00
1- 1 , I 0.80 0.84 0 88 0 92 0 96
0 80 0.81 0.82 0.83 t ,s
t,S Simulation resultsfor Fig. 1: is
.5 SP WM technique
Inverter output Vd
Operation frequency Proposed inverter Inverter in Fig. 2 Proposed inverter with PFC
f, Hz THD,(%) PF,(%) RF,(%) THD,(%) PF,(%) RF,(%) TWD,(%) PF,(%) RF,(%)
60 120 67.1 1.21 108 71.1 0.89 9.8 99.3 0.97
45 125 60.3 1.91 117 61.3 0.79 11.8 98.7 1.45
30 '1 55 52.6 2.71 134 58.7 0.52 16.2 98.4 1.94
460 IEE Proc -Electr Power Appl, Val 143, No 6, November 1996
3.2 Control of the inverter
l6Ol The inverter generates two-phase, variable-frequency,
variable-voltages to the primary and auxiliary windings
of the motor. The constant Vlf strategy is used to
determine the output voltage magnitude based on the
input speed (i.e. frequency) command. The two-phase
0.80 0.84 0.88 0.92 0 96 voltage output waveforms have a phase difference of
t.s 90" and are modulated using the sinusoidal pulse width
Fig.8 Simulation resultsfor Fig. 1: Vci modulation (SPWM). Based on the symmetric
155-
component concept, the condition for the motor to
have zero torque pulsation is to operate the motor with
two-phase excitation having a 90" phase relation and
1 the current amplitude in inverse ratio to the machine
zL1501
2.
145
turns [6]. A further approximation to achieve the
condition of zero torque pulsation is to generate two-
phase voltages having a 90" phase difference and the
0 80 0.84 0.88 0.92 0.96 primary-to-auxiliary voltage-magnitude-ratio is
t.s inversely proportional to the machine turns [6]. Notice
Fig.9 Simulation resultsfor Fig.2: V,
that, for the unsymmetrical two-phase induction motor,
3 Modelling, control and simulation the auxiliary winding has higher voltage level than the
primary winding and, consequently, the primary
3.1 Motor dynamic equations winding has higher current level. This type of operation
The state-space dynamic equations for a two-pole, two- is different from a conventional motor which normally
phase, unsymmetrical induction motor are formulated has balanced two-phase currents. However, the
in the stationary reference frame as [14] proposed approach results in less torque pulsation and
noise. Thus, to apply the proposed inverter to a
conventional motor without altering its original
construction and winding arrangement, the voltage and
current ratings of the motor windings must be carefully
evaluated.
The same inverter can also be applied to a
symmetrical two-phase induction motor with better
performance. Both windings of the motor have the
same level of voltages having a 90" phase difference.
The resultant phase currents are also balanced and a
where the subscripts s, S, r and R denote the smooth generated torque is obtained. This is then
unsymmetrical stator and rotor windings. For the identical to the case of a balanced three-phase
symmetrical case, X,, = Xss and others similarly apply. induction motor driven by a three-phase inverter.
For starting purpose, a capacitor is usually connected
in series with the auxiliary winding of the motor. The 3.3 Computer simulation
term P / O b x , , given in eqn. 1 is replaced by p / O b x , + In many applications employing the two-phase
ob/pXc. Let z = i&/p, which can be augmented with induction motors, the loads usually exhibit nonlinear
eqn. 1 to form a complete set of equations for characteristics, such as the rolling-piston type
describing the motor dynamics with a starting compressor commonly found in a room air conditioner.
capacitor. The generated motor torque is However, the dynamic equations of these nonlinear
loads contain many design parameters which are not
easily obtainable from the manufacturers. The
simulation results presented in this section are all based
and the electromechanical equation is on the constant load torque assumption. Still these
2 results provide an assessment on the effect of the
J-pwT
P
+
Bw, TL = Te + (3) inverter in reducing the motor torque pulsation.
The dynamic equations of the unsymmetrical two- To evaluate in-depth the various characteristics of an
phase induction motor are more complicated than inverter-driven motor drive, several operational modes
those of the symmetrical three-phase induction motor. are chosen for a comparison, which all use the same
The unsymmetrical winding arrangement in the two- induction motor with unsymmetrical two-phase wind-
phase induction motor causes greater torque and speed ings. The motor parameters are tabulated in Table 2
fluctuations. However, both the symmetrical two-phase and the ratings are IlOV, 6OH2, 1/4HP, and four-pole.
and three-phase induction motors have similar In all simulated cases, the load torque is fixed at 1N m.
performance. The operational modes are described in the
Table 2: Induction motor parameters
XcS = 14.5 P
170 172 174 1 76 178 180 Fig. 13 Simulated average speeds for cases 1-5
t.s
-+- case 1, capacitor start
---case 2, capacitor run (CR = 183pF)
Fig. 10 Simulatedspeed responsefor case 3
--
-*- case 3, capacitor run (CR = 15.4pF)
-x-
case 4, two-phase (Sine source)
case 5, two-phase (inverter)
14407
1425 1,
1 70 172 174 176 1 78 180
t.5
Fig. 11 Simulated speed vesporisefor case 4
C
1435 1I
1
,
,
1
I I
I I
I I
I 1
14201 , I I , , , , , 35 40 45
f. Hz
50 55 60
170 1 72 1.74 176 1.78 180 Fig. 14 Simulated average slips for cases 1-5
t.s -t- case 1, capacitor start
Fig. 12 ,%udatedspeed responsefor case 5 ---case 2, capacitor run (CR = 183 pF)
-*- case 3, capacitor run (CR = 15.4pF)
-0- case 4, two-phase (Sine source)
-x- case 5 , two-phase (inverter)
For cases 4 and 5, the ratio of primary-to-auxiliary
voltage magnitude is set equal to NJ/NS The input The torque variation, which is defined as the differ-
frequencies are varied from 35 to 60Hz with an ence between the peak (highest) to peak (lowest) torque
increment of 5Hz. F o r f = 50Hz, the speed responses values in a steady-state torque response, is computed
for cases 3, 4, and 5 are shown in Figs. 10, 11 and 12. for each case and is summarised in Fig. 17. Again, the
The average speed, wave, defined as the mean of the torque variation in case 5 is almost identical to that of
peak and lowest speeds of the steady-state speed case 4 and is the smallest as compared to cases 1-3.
response, is calculated for each case and is shown in Finally, the steady-state generated motor torque and
Fig. 13. Define the synchronous speed we = 2zf and the load torque responses for cases 3, 4, and 5 are shown
slip wSl as in Figs. 18, 19 and 20. The proposed approach, case 5,
shown in Fig. 20 reduces the generated torque pulsa-
tion quite effectively as compared to the conventional
462 IEE Proc -Elect7 Power A p p l , Vol 143, No 6,November 1996
approach, case 3. In fact, the responses of case 5 and experimental system is shown in Fig. 21. The motor
case 4 are almost identical. used in the experiments has unsymmetrical two-phase
windings with removable start and run capacitors and
centrifugal switches according to the experimental
needs. The primary-to-auxiliary turn ratio is M.3. The
ratings of the motor are llOV, 60Hz, four-pole, 1/2
HP. The motor is coupled to a 400W DC generator,
which, by adjusting its field voltage, provides variable
load torque to the motor. To measured the torque
responses, a torque sensor is mounted on the same
shaft which couples the induction motor and DC
generator. The shaft speed is also measured using a
tachometer. To further record the effect of torque
pulsation, a portable type accelerometer which can be
magnetically attached to the motor frame is used. The
0 0 accelerometer measures the mechanical vibration and
35 40 45 50 55 60 sends the signal to the HP35665A signal analyser for
f,Hz
Fig. 15 Simulated peak-to-peak speed variationsfor cases 1-5 spectrum analysis. The vibration spectrum provides an
-._
-+-
-*-
case 1, capacitor start
case 2 , capacitor run (CR = 183 pF)
case 3, capacitor run (CR = 15.4 pF)
evaluation of torque pulsation effect. A microphone is
also installed at a distance of 3cm from the motor. The
-0- case 4, two-phase (Sine source)
-x- case 5, two-phase (inverter) noise signals generated from the drive system are sent
to the signal analyser for spectrum analysis.
0
2
E
Z
d
' 1
e
c
Q
c
.4 0
C
E 1.70 1.72 1.?4 1.76 1.78 100
t.5
a Fig. 18 Simulated torque response for case 3, 50Hz
21
35 40 45 50 55 60
f.Hz
Fig. 16 Simulatedprimary stator line currentsfor cases 1-5
-+-
-.-
_ *_
case 1, capacitor start
case 2, capacitor run (CR = 183 pF)
case 3, capacitor run (CR = 15.4 pF) 1.70 1.72 1.74 1.76 1.78 1.80
-e case 4, two-phase (Sine source) t. 5
-x- case 5, two-phase (inverter) Fig. 19 Simulated torque response for case 4, 50Hz
E
z
d 1
P
c
I I
1
I I
I I
I I
1.70 1.72 1.7L 1.76 1.78 1.80
t.5
Fig.20 Simulated torque response for case 5, 50Hz
-
v
-
" "
e " L
o r , 1 I
i l r l i
35 40 45 50 55 60
f,Hz
Fig. 17 Simulated torque variationsfor cases 1-5
-.-
-+-
-*-
case 1, capacitor start
case 2, capacitor run (CR = 183 pF)
case 3, capacitor run (CR = 15.4 pF) driving HP35665A torque -
-0- case 4, two-phase (Sine source) signal
-x- case 5, two-phase (inverter) circuit ana,yser meter
6
The inverter consists of a half-bridge rectifier, two
series capacitors C1 = C2 = 2000pF and two-pairs of
IGBTs. The inverter provides two-phase, variable-
frequency, vanable-voltages to the primary and
auxiliary windings of the motor. The control of IGBT
switching is determined by a single-chip E-
E
U
-*-
case 1, capacitor start
case 2, capacitor run (CR = 75p.F)
voltage phasor is d2 x 110 L O"/A,where A = 1.3 is the -&
case 3, capacitor run (CR = 2 0 p )
case 4, two-phase (Sine source)
turn ratio and the auxiliary voltage phasor is d2 x 110 -x- case 5, two-phase (inverter)
L 90".
1
0
35 40 45 50 55
f , Hz
Fig.25 Measuredprimary stator currents for cases 1-5
-.-
-+-
-*_
case 1, capacitor start
case 2,
case 3,
capacitor run (CR = 75pF)
capacitor run (CR = 20p.F)
-0- case 4, two-phase (Sine source)
-x- case 5 , two-phase (inverter)
35 40 45 50 55 60
f , Hz
Fig.26 Meamred vibration signal levels
-.-
-+-
*
_-
case 1, capacitor start
case 2, capacitor run (CR = 75wF)
case 3, capacitor run (CR = 2 0 s )
... . . . . ... . . . . ... . . . .
. . .
-0 case 4, two-phase (Sine source)
-x- case 5, two-phase (inverter) . . .
.. ., ..
.............. . . . .. . . . . . . . . . .. . . . . . .
-40 7 .. .. .. .. .
. .
. .
.
. . . ...... . .
I I
. . . . .
35 40 45 50 55 60 . . . . .
f, Hz
Fig.27 Measured noise signal levels
-.-
-+-
-*-
case 1, capacitor start
case 2, capacitor run (CR = 7 5 p)
case 3, capacitor run (CR = 20m) t
-0 case 4, two-phase (Sine source) .
.
. . .
. . . . .. . . . .. . .
-x- case 5 , two-phase (inverter)
.. .
.
.
.
.
.
Fig.30 Measured speed ripple responsesfor cases 3-5 at SO Hz
. . . . . . . . (i) Case 3
(ii) Case 4
(iii) Case 5
5
t
-60
(ii) -80
0 100 200 300 400
f , Hz
. . . .. .. . . Fig.31 Memured spectrum of noise signal: Case 3
.. .. .. ..
IEE Proc-Elect?. Power Appl., Vol. 143, No. 6, November 1996 465
With the rapid advancement in microelectronics tech- 4 HOLMES, D.G., and KOTSOPOULOS, A.: ‘Variable speed con-
trol of single and two phase induction motors using a three phase
nology, the design of a control-and-power application voltage source inverter’. IEEE Industrial Applications Society
specific IC based on the architecture presented in this annual meeting conference records, 1993, pp. 613-620
paper is feasible. By combining the control-and-power 5 ALEXA. D.: ‘Static frequency converter for supplying an asyn-
chips with the current small-power two-phase induction chronous two-phase motor’, ZEE Proc. B, 134, (I), pp. 57-60
6 COLLINS, E.R., PUTTGEN, H.B., and SAYLE, W.E.: ‘Single-
motors, the drive performance can be greatly phase induction motor adjustable speed drive: direct phase angle
improved. control of the auxiliary winding supply’. IEEE Industrial Applica-
tions Society annual meeting conference records, 1988, pp. 246-
252
6 Acknowledgments 7 COLLINS. E.R.: ‘Torque and slip behavior of single-phase
induction motors driven from variable-frequency supplies’, IEEE
Trans. Ind. Appl., 1992, 28, (3), pp. 710-715
This research was partially supported by the National 8 ZHU, Z.Q., and HOWE, D.: ‘Vibrational torques in single-phase
Science Council, Taiwan, under grants NSC82-0416-E- induction motors and their relation to vibration and noise, Elec-
01 1-096 and NSC83-0416-E-011-001. tric Mach. Power Syst., 1992, 20, pp. 483492
9 YOUNG, C.-M., LIU, C.-C., and LIU, C.-H.: ‘Vibration analy-
References sis of rolling piston-type compressors driven by single-phase
induction motors’. IEEE international conference on Industrial
electronics, control and instrumentation, 1993, pp. 918-923
MHANGO, L.M.C., and CREIGHTON, G.K.: ‘Novel two- 10 NISHIHARA, M.: ‘Power electronics diversity’. Proceedings of
phase inverter-fed induction-motor drive’, IEE Proc. B, 1984, 131,
(31, pp. 99-104 1990 international Power electronics conference, 1990, pp. 21-28
JANG, D., CHA, G., KIM, D., and WON, J.: ‘Phase-difference 11 SEDRA, A.S., and SMITH, K.C.: ‘Microelectronic circuits’ (CBS
control of 2-phase inverter-fed induction motor’. IEEE Industrial College Publishing, 1987)
Applications Society annual meeting conference records, 1989, 12 RASHID, M.H.: ‘Power electronics: circuits, devices, and appli-
pp. 571-578 cations’ (Prentice-Hall, 1993)
JANG, D.-H., and WON, J.-S.: ‘Voltage, frequency, and phase- 13 BOYS, J.T., and GREEN, A.W.: ‘Current-forced single-phase
difference angle control of PWM inverters-fed two-phase induc- reversible rectifier’, ZEE PYOC.B, 1989, 136, (9,pp. 205-211
tion motors’, ZEEE Trans. Power Electron., 1994, 9, (4), pp. 377- 14 KRAUSE, P.C.: ‘Analysis of Electric Machinery’ (McGraw Hill,
383 New York, 1987)
466 IEE Proc.-Electr. Power AppL, Vol. 143, No. 6, November 1996