Documente Academic
Documente Profesional
Documente Cultură
by
Me: Li
A project
presented to Ryerson University
in partial fulfillm ent o f the
requirements for the degree o f
M a ste r o f E n g in eerin g
in the Program o f
Electrical and Computer Engineering
M ei Li 2005
PROPERTYOF
RYERWN WimtTY UBRARV
UMI
UMI Microform EC53041
Copyright 200 8 by ProQuest LLC
All rights reserved. This microform edition is protected against
unauthorized copying under Title 17, United States Code.
ProQuest LLC
789 East Eisenhower Parkway
P.O. Box 1346
Ann Arbor, Ml 481 06-1346
Instructions on Borrowers
Ryerson U niversiry requires the signatures o f all persons using or photocopying this
sis. Please sign below, and give address and date.
Abstract
This thesis presents the introduction, analysis and experimental verification of the sixpulse SCR rectifier and multi-pulse SCR rectifiers.
In this thesis, the six-, 12-, 18- and 24-pulse SCR rectifiers with inductive and
capacitive loads are introduced. The line current THD and the input PF of various
rectifiers are investigated. The principle of the harmonic elimination through phaseshifting transforms is analyzed by Fourier analysis and positive/negative sequence
analysis. The experimental verification is accomplished on a prototype of the 12-pulse
SCR rectifier.
Key words; multi-pulse SCR rectifier, harmonic elimination, line current THD.
IV
Acknowledgements
Thanks also go to all professors in the pow er field and research personnel in the
moratory for Electric Drive Research and Applications (LEDAR) at Ryerson
University, w ho offered valuable help on my study.
Table of Contents
Author s Declaration
ii
Instructions on Borrowers
iii
Abstract
iv
Acknowledgements
Table of Contents
vi
List of Figures
ix
List of Acronyms
xi
Chapter 1 Introduction
...........................................................................14
16
17
VI
18
S u m m ary ......................................................................................................................... 20
a p te r 3 12-PuIse S C R R ectifier
21
21
31
34
C h a p te r 4 18-Pulse SC R R ectifier
37
37
YU
44
45
45
53
Chapter 7 Conclusions
65
67
Appendix B THD and PF Curves of the Six- and Multi-pulse SCR Rectifiers 77
References
79
vm
List of Figures
Load voltage waveform s o f the six-pulse SCR rectifier with inductive loads
at Ls=0.
2-4 Current w aveform s o f the six-pulse SCR rectifier with inductive loads at
11
14
15
2-7 THD and PF curves o f the six-pulse SCR rectifier with inductive loads.
17
2-8 Circuit configuration o f the six-pulse SCR rectifier with a capacitive load.
17
18
2-10 W aveform and harm onic spectrum o f lai = lp u and lai =0.2pu (six-pulse).
19
2-11 THD and PF curves o f the six-pulse SCR rectifier with capacitive loads.
20
22
3-2 Voltage and current waveform s o f the 12-pulse SCR rectifier with inductive
loads.
24
3-3 Calculate the average load voltage o f the 12-pulse SCR rectifier with
inductive loads.
25
3-4 Find the phase displacements o f harmonics in the delta-conneeted winding line
current when the transform er is -30 phase shifted.
25
28
29
IX
30
31
3-9 THD and PF curves o f the 12-pulse SCR rectifier with inductive loads.
32
32
3-11 Current waveforms of the 12-pulse SCR rectifier with capacitive loads.
33
3-12 Waveform and harmonic spectrum of lai = lpu and lai =0.2pu (12-pulse).
34
3-13 THD and PF curves of the 12-pulse SCR rectifier with capacitive loads.
35
38
39
40
4-4 THD and PF curves of the 18-pulse SCR rectifier with inductive loads.
41
41
42
4-7 Waveform and harmonic spectrum of lai = lpu and lai =0.2pu (18-pulse).
43
4-8 THD and PF curves of the 18-pulse SCR rectifier with capacitive loads.
44
46
47
48
5-4 THD and PF curves of the 24-pulse SCR rectifier with inductive loads.
49
49
50
5-7 Waveform and harmonic spectrum of lai = lpu and Igi =0.2pu (24-pulse).
51
5-8 THD and PF curves of the 24-pulse SCR rectifier with capacitive loads.
52
53
55
59
60
60
63
List of Acronyms
distortion factor
F displacement pow er factor
O - gate turn-off thyristor
3T insulated gate bipolar transistor
__ /L - KirchhofP s voltage law
PF pow er factor
RM S - root m ean square
SCR - silicon controlled rectifier
TDD - total dem and distortion
THD - total harm onic distortion
XI
Chapter 1 Introduction
IEEE Standard 519-1992 defines harmonic limits for electrical pow er systems. Table
1-1 is extracted from the standard, which is applicable to six-pulse rectifier loads. For
m ulti-pulse rectifier loads, the limits are increased by a factor o f ^fq /S under certain
conditions [1], in which q is the pulse number o f the rectifier. However, since only ac
voltages are available from the utility supply, rectifier equipment could be considered as
dc pow er generator, converting ac to dc; hence based on Table 1-1, the Total Demand
D istortion (TDD) generated by them should be limited within 5%.
<11
ll< h < 1 7
17<h<23
23<h<35
35<h
<20*
4.0
2.0
1.5
0.6
0.3
20<50
7.0
3.5
2.5
1.0
0.5
8.0
50<100
lO.O
4.5
4.0
1.5
0.7
12.0
100<1000
12.0
5.5
5.0
2.0
1.0
15.0
>1000
15.0
7.0
6.0
2.5
1.4
20.0
To meet the restriction of IEEE Standard 519-1992, many methods could be applied
to eliminate or reduce harmonic currents generated by rectifiers. For example,
(i) Using passive filters. Line reactors (inductors) are often used in conjunction with
capacitors in a rectifier to filter specific harmonic currents. However, the cormection
o f capacitors could cause resonance conditions that can magnify harmonic current at
certain frequency to a harmful level [6-7].
(ii) Using certain switching techniques. Switching techniques, for instance the pulse
width modulation technique, could be used to eliminate harmonic currents. In high
power applications, the premise for using these techniques is that the switching
devices must be the gate-turn-off type, such as GTOs or IGBTs [8-9].
(iii)Using multi-pulse rectifiers. Multi-pulse rectifiers are designed based on phaseshifting transformers, which enable certain harmonics to be cancelled from the
rectifiers line currents. This method is especially practical for harmonic elimination
in high power applications [1-3,9,10].
This thesis provides full analysis o f multi-pulse SCR rectifiers for engineering
reference, because they are widely used in industry and play an important role in
harm onic eliminations.
This chapter begins with the introduction o f the six-pulse SCR rectifier, and then
steps into m ulti-pulse SCR rectifiers, which overcome shortcomings o f the six-pulse SCR
rectifier. The objective o f the thesis is also presented. At the end o f the chapter, the
outline o f the thesis is provided.
F ig. 1-1 Configuration o f the six-pulse SCR rectifier with two devices connected in series.
In the six-pulse SCR rectifier, it might be required to connect SCRs in series in large
horsepow er applications to block high voltages, under which a single SCR caimot handle.
V oltage sharing is the main problem o f the com ponent in-series connection. Fig. 1-1
shows the configuration o f the six-pulse SCR rectifier with two devices connected in
series. The two SCRs, for instance SCRja and SCRjb, should be turned on (or turned off)
sim ultaneously. I f not, the one turned on later (or turned o ff earlier) will undertake a
higher voltage and thus may be damaged. If more SCRs are required to be in series in
high-voltage applications, the voltage-sharing problem becomes more critical, because
the synchronized switching is more difficult to realize.
Fig. 1-2 shows the simplified diagram of the 12-pulse SCR rectifier, which consists of
a phase-shifting transformer and two six-pulse SCR bridges that are connected in series
on the dc side. Fig. l-2(a) shows the inductive load condition, for example load
commutated current source inverter powered motor drives or high-voltage direct-current
transmission systems. Fig. l-2(b) shows the capacitive load condition, such as high power
battery chargers or dc motor drives.
Fig. 1-2
The main feature o f the 12-pulse SCR rectifier is that two SCR bridges are connected
in series rather than two SCRs, thus the voltage sharing is avoided. As a result, the system
becomes more reliable. Another advantage of the 12-pulse SCR rectifier is that the line
current THD is reduced substantially owing to the use o f the phase-shifting transformer.
The details o f the harmonic elimination will be analyzed in Chapter 3.
For the 18-pulse SCR rectifier, a phase-shifting transformer with three secondary
windings is connected to three six-pulse SCR bridges. Similarly, connecting a phaseshifting transform er with four secondary windings to four six-pulse bridges introduces the
24-pulse rectifier. Since more SCR bridges are connected in series to tolerate high
voltages, and more low-order harmonics are eliminated from the line current, the 18- and
24-pulse SCR rectifiers are widely used in high power applications. For example, they
could be applied as front ends o f high power motor drive systems.
To use com puter simulation to investigate the performance o f the six-pulse SCR
rectifier and multi-pulse SCR rectifiers with inductive and capacitive loads.
To com pare the performance o f the six-pulse rectifier and multi-pulse rectifiers by
providing voltage and current waveforms under different load conditions.
To
analyze
the
harmonic
elimination
theory
by
Fourier
analysis
and
To obtain the line current THD and the input PF under different operating
conditions by simulations.
Chapter 2 presents the performance of the six-pulse SCR rectifier with inductive and
capacitive loads, respectively. The load voltage and current, the RMS and THD of the
line current and the input PF of the rectifier under ideal and non-ideal conditions are
discussed.
Chapter 3 presents the performance of the 12-pulse SCR rectifier. The harmonic
elimination theory is illustrated by Fourier analysis and positive/negative sequence
analysis, respectively. Voltage and current waveforms, harmonic spectrum and THD of
the line current as well as the input PF under different operating conditions are provided.
Chapter 4 and 5 presents the performance of the 18- and 24-pulse SCR rectifiers,
respectively. Typical voltage and current waveforms, harmonic spectrum and THD of the
line current, the harmonic elimination theory and the input PF are discussed.
This chapter presents the performance o f the six-pulse SCR rectifier with inductive
and capacitive loads. Voltage and current waveforms are provided to show the operation
o f the rectifier. For the inductive load condition, the average load voltage and current
under ideal and non-ideal circuit conditions are derived; the RMS and THD o f the line
current are analyzed; the input PF is calculated. For the capacitive load condition, the line
current THD and the input PF are discussed.
-V+
)- +
F ig. 2-1
Fig. 2-1 shows the circuit configuration of the six-pulse SCR rectifier with an
inductive load. The input of the SCR rectifier is a balanced three-phase power supply,
which consists o f the following phase voltages:
v = -j2Vm sin(),
= 'l 2 V, sin 0 )t~-
(2.1)
In
(2.2)
T,
An
(2.3)
S in
7t
(2.5)
cot--
(24)
(2.6)
Vll
Ls represents the total line inductance between the power supply grid and the rectifier
bridge, which includes the equivalent inductance of the power supply and the inductance
o f a three-phase reactor that is often added to the system to reduce the line current THD.
L l is the dc choke, which is used to reduce the ripple component of the load current.
For easy o f analysis, assume that the value of L l is infinite, and then the inductive load
can be replaced by a dc current source Ii as shown in Fig. 2-1.
ab
a=
a = 2t i /3
"TV,
0
a =1112)
a= SnJ6
a =n
Fig. 2 -2
Load voltage waveform s o f the six-pulse SCR rectifier with inductive loads at Z^=0.
F ig. 2-3
7 t/6
(7 i/6 )+ a
{ntiyi-a
From Fig. 2-2, it can be seen that the dc voltage obtained from the six-pulse SCR
rectifier contains six ripples in one supply voltage cycle. Based on Fig. 2-3, we could
have
areaAj _
;r/3
=
3 V2
1 r/2 +a
;r/3^/'
(2.7)
From equation (2.7), when the delay angle a is less than 90, the average load voltage
is positive. The energy is transferred from the source to the load. In this condition, the
SCR rectifier is called operating in the rectification mode. When the delay angle a is
greater than 90, the average load voltage becomes negative. The energy is transferred
from the load to the source. The rectifier is called working in the inverting mode.
Fig. 2-4 shows the current waveforms of the six-pulse SCR rectifier. iscRi, hcR3 and
isaware the currents passing through SC R i, SCR3 and SCR5, each of which conducts for
27t/3 radians in one supply voltage cycle in turn. The similar condition happens to
SCR2, SCR4 and SCRs as well, because at any moment there must be two SCRs
conducting, one (SCR], SCR3 or SCR5) connected to the positive pole of the rectifier
bridge, and the other (SCR2, SCR4 or SCRs) to the negative pole, to let the current flow.
The waveform of zs'o?; and
are not shown in Fig. 2-4, but they can be easily obtained
by referencing to the iscR4 waveform. The line current of phase A can be calculated by 4=
is c R i- is c R 4 -
Thus /a consists of a positive part and a negative part, which are contributed by
SCR] and SCR4, respectively. / is the load current, which is the sum of iscRi, kcRs and
(2.8)
~R
h equals to h,avg-
10
^SCRI
>
^SCR3
0
sCRS
0
I,.
^SCR-l
. 2rt/3 .
li d 'i
jc/3
4 ti/3
2k
571/3
7T
'
F ig. 2 -4
The RM S value o f the dc load current equals to its average value, i.e.
TL ,R M S = TL ,avg
'
(2.9)
^^
(2. 10)
Since SCRs in each phase conduct one-third o f one supply voltage cycle ( 2 ;r /3
radians), w e have
T
S C R .a v g
L,avg
g
11
(2. 11)
2) Line Current
RMS Value of the Line Current
Based on Fig. 2-4, the RMS value of the line current can be calculated by
h,Rus - ^ 2 ^
= J | / , ^ = 0.816/,,
Hence once the amplitude of the dc load current is known, the RMS value of the line
current can be calculated.
In Fig. 2-4, ia is shown as an even function with period T - I tt. Thus it can be
expressed by the following Fourier Series:
M
/7
00
yiTT
n
00
= - f + Z a , co s- Alt =
+ E a , c o s n a t,
z
=1
L IZ
(2.13)
rt=i
frn. /
\
UK
. ,r
\
2
(^'^^/(i/)cos------ }td{)t) = f i^{)t)cosnJtd{)t)
r/2 " '
'
r/2
' ' ;r
cosnW((uf)+
_2/.
nn
( n K \ . ( ln K \
+ sm ----I 3 J
L 3
(2.14)
( = 0,1,2,...)
the phase angle of iai with respect to v^, we
have:
4=1. l/i[cos(ffir - ^ ) - - cos5(ffiT - ^ ) + - cos7(<af - ^ )
1
^ I
^
cosl l ( u /- ^ ) + cosl3(<rf-^)-...]
12
(2.15)
It can be noticed that there is no triplen harmonics (3^**, 6^'', 9^, 1 2 ^ ...) in ia, because the
rectifier is a balanced three-phase system.
Z ^Ia c o sn ( X -(^),
(2.16)
0= 1, 5, 7,11
w here Ian is the RM S value o f each harmonic com ponent in ia, then the RM S value o f ia
can be expressed as:
^ a ,R M S J
(2-17)
V = 1. 5. 7.11
-.... ,
(2.18)
^ a \,R M S
w here Iai,RMs is the RMS value o f iai, the fundamental component o f ia- From equation
(2.15), the am plitude o f iai can be obtained by:
7.i,m = 1 -1 7 ,,
(2.19)
7 i . i . . = % = 0 .7 8 /,.
(2.20)
then
J ( 0 . 8 1 6 / , j - - ( 0 .7 8 0 1 ^ ^ 3 ^ ,
0.78/,_
(2.21)
3) I n p u t P F o f th e S C R R ectifier
In a balanced three-phase system, the power factor o f the system is equivalent to that o f
each phase. Per-phase average (real) pow er can be obtained from the phase voltage and
the fundam ental com ponent o f the phase current, i.e.:
^
cos
(2.22)
2 tv
Per-phase apparent pow er is obtained from the phase voltage and the total phase current,
i.e.:
13
(223)
Then the power factor is:
PF = - =
= ^ c o s ^ i =DFxDPF.
(2.24)
where DF is the distortion factor that is defined as: DF = 4 , / 7^, and DPF is the
displacement power factor that is defined as: DPF = cos^,. In the SCR rectifier circuit,
is the result of the delay angle a. Under ideal circuit condition,
cos <fi\ = cos . From THD =
PF^
; hence
DPF
cos a
Vl + 777D"
Vl + 777D"
= 0.955 c o s .
(225)
In Fig. 2-6(a),
due to the Ls, the current in SCRi reduces from It to zero gently and the current in SCR3
increases from zero to h gradually. Therefore, there exists a short commutation interval y.
SCR, I'^SCR
Fig. 2-5
14
> M
-4
V
>
cat
(b)
F ig. 2 -6
D uring the com m utation interval y, as S C R i and SCR3 are both on, the source is shortcircuited (indicated by the dotted line in Fig. 2-5). Therefore, there w ould be an output
voltage loss, w hich is indicated by the shaded area in Fig. 2-6(b). Based on the
K irchhofPs V oltage Law (KVL) at node P in Fig. 2-5, we have
=v,-L.
dL
dt
dhb
dt
(2.26)
diu
dt
(2.27)
From which:
di
dt
Since
dig
dt
. , then
dt
(2.28)
dt
dt
Hence
v + v.
V/. =
(2.29)
15
4 = r 4 f - d W = f . ' i A = ffliA -
(2-31)
=^ ^ ^ .
n il)
(2.32)
7t
From equation (2.32), we know that the average dc voltage loss is proportional to Ls and
h . The corrected average load voltage VL,avg_c and current h,avg_csre:
(2.33)
(2.34)
y 4^^)
(235)
Y = cos
-1
rz
V2)I
c o s a -----------7^ cc.
(2.36)
Fig. 2-1 {c) shows the PF curves. From PF = DPF I -i\ + THD^ , we know that the PF
is related to THD and DPF. At any fixed a , when the load current II decreases, the RMS
16
value o f ia decreases. As a result, the phase displacement <j>i decreases, the DPF (=cos^/)
increases, and the PF increases.
THD o f/ at a= 0
-------1---------
Ou.
28
1 30
26
29
A
B
0.8
0.6
24
28
A: Lj=0.05 p.u.
22 B: Lj=0.10 p u.
C: Lj=0.15 p.u.
2
0
0.2
0.4
C: a=60
B: a=40
A: a=20
27
26
0.6
0.8
X,
0.2
0.2
0.4
0.6
0.8
hi (P u )
(a)
F ig. 2 -7
A: a=0
B: (i=20
C: a=40
04
0.2
0.4
0.6
(b)
0.8
(c)
'nSCJif A5CJJ
C=x>
F ig . 2-8
17
h, (P-U-)
hi (P )
vll is
vll=E;
then it begins to
decrease till reaches zero. As a result, two humps appear in the positive half cycle of 4The same phenomenon happens to the negative half cycle of ia as well. The load current
z'i is the summation of the positive half cycle of the three line currents.
W
n /Y v w w \j
Fig. 2-9
The value of E varies slightly with the loading conditions. Under light load
conditions, E is close to the peak of vll, therefore the load current ii could be
discontinuous. As the.load becomes heavier, a bigger current will be drawn, which causes
a larger voltage drop across
becomes smoother. When the load increases to a certain level, the load current becomes
continuous that indicates a continuous current operation mode of the SCR rectifier.
18
40
4/=l p.u.
TH0=34.86 %
>k
1.414
0
0.5
5th
7th 11th 13th 17th 19th
-40
0
i^atl^i= 0.2
10
15
20
30
25
Frequency (p.u.)
10
THD=72.93 %
5th
0.282
0
0.5
7th
11th 13th 17th 19th
-10
F ig. 2-10
10
15
20
30
25
Frequency (p.u.)
Fig. 2-11 (a) and (b) show the THD curves at different value o f Ls, lai and a . It can be
seen that the THD values are higher under lighter load conditions. Fig. 2 -1 1(c) shows the
PF curves under different Ls values. It can be seen that a larger Ls increases the input PF
under lighter load conditions, but the PF could not be im proved m uch under rated load.
19
glOO
^ 80
\
.\
\ \
60
40
20
0.2
140
A; Lj=0.05p.u.
Q
.
Z 120
B; Lj=0.10 p.u.
C: Lj=0.15 p.u.
100
\A
80
60
40
c
20
0.4
0.6
0.8
C:a=60
B: a=40
A: a=20
--- .A
L(pu)
L (pu.)
Fig. 2-11
0.90.8
0.7
0.6
fc
. 1
CU
A: Lj=0.05 p.u.
B; Lj=0.10p.u.
C: Lj=0.15 p.u.
0.2
0.4
0.6
0.8
THD and PF curves o f the six-pulse SCR rectifier with capacitive loads.
2.3 Summary
This chapter presents the performances of the six-pulse SCR rectifier with inductive
and capacitive loads. The following conclusions can be made:
For the inductive load condition, the average load voltage is determined by the RMS
value of the line-to-line supply voltage, the delay angle a, and the line inductance Ls.
The Ls causes the commutation interval, which impacts the line current THD. The
input PF is a function o f the THD and the DPF.
For the capacitive load condition, the line inductance Ls and the load current
amplitude influence the THD and the DPF of the line current. The input PF can be
calculated based on the THD and the DPF.
The line current THD of the six-pulse SCR rectifier with either the inductive or the
capacitive load cannot meet the harmonic current requirement defined by IEEE
Standard 519-1992.
20
4,(pu)
As discussed in Chapter 2, the six-pulse SCR rectifier has a high line current THD,
w hich does not m eet the IEEE Standard 519-1992. Using m ulti-pulse (12-, 18- or 24pulse) SCR rectifiers, in which a phase-shifting transfom ier and a couple o f six-pulse
SCR rectifiers (at least two) are applied, could eliminate or decrease low-order harmonics,
so as to reduce the line current THD.
This chapter presents the performance o f the 12-pulse SCR rectifier w ith inductive
and capacitive loads. The voltage and current waveforms are illustrated. Based on Fourier
analysis and positive/negative sequence analysis, the principle o f the harmonic
elim ination through the phase-shifting transform er is discussed. The line current THD
and the input PF under different operating conditions are obtained by computer
sim ulations.
21
bsl
5 = 0
csl
5 = -30
- t-
Am
(MTYVA.
3,
a s l
li
V ,,
5 = 0
0 _ ^ W Y V \ ^as2
5 = -30
it
the phase-shifting transformer, and Lik stands for the total transformer leakage inductance,
which is the summation o f the primary windings leakage inductance reflected to the
secondary side and the secondary windings leakage inductance.
(3 1)
= 2,
(32)
V s'
a = z % u ,- /r w = o .
22
(13)
(3.4)
(3.5)
where
and V^^bi ^re RMS line-to-line voltages o f windings Tp, Tsi and Ts2 ,
^ s \ aiid N ^2
the num ber o f turns in each phase o f Tp , Tsi and Ts2 ', S is the phase
connected, there is no phase shift between their line-to-line voltages. Ts2 is delta-connected,
and thus its line-to-line voltages are -30 phase shifted with respect to the line-to-line
voltages o f Tp.
= 0 and
Fig. 3-2(c) to Fig. 3-2(f) show the current waveforms, where a, iasi and ias2 are line
currents o f w indings Tp, Tsi and Ts2 ', iasi and
transform er prim ary side; ii is the load current with amplitude Z.; iscR ij and iscR2 j are
SCRx currents o f the two SCR bridges, respectively. Fig.
transform er secondary side, iasi and ias2 are identical, but with a 30 phase shift. On the
transform er prim ary side, iasi and w 'a r e in phase with Z, and their summation is Z. On
both sides o f the transformer, iasi and iasi are identical in shape, but have different
am plitude due to the turns ratio; ias2 and w ' have different shapes because o f the 30
phase shift.
23
1.077/,
0.7885/,
0-577/i ^
0.5/,
0.2885/,
Tt/6 w3 n/2
I it o / ^/
Fig. 3-2
Voltage and current waveforms o f the 12-pulse SCR rectifier with inductive loads.
and 4,2.ovg
= 0 ), we have
cosor,
(3.6)
shows the phasor diagram, in which Vi is the phasor summation of Vu and Vu- Thus
^L.avg_i2 the average value of vi, can be calculated as follows:
=K| =K +
= 1.93 X ( l . 3 5 4 , , M co sa).
24
(3.7)
F ig. 3 -3
Calculate the average load voltage o f the 12-pulse SCR rectifier w ith inductive loads.
(3.8)
= ^0.62161^ = 0.7884/^
Com pared to equation (2.12), this RMS value is lower than that o f the six-pulse rectifier
due to the harm onic elim ination in the line current.
N,
bc_ph
ha ph
F ig. 3 -4
^2a
25
Well use the circuit in Fig. 3-4 to find the phase displacements of harmonics in the
delta-connected winding line current when the transformer is -30 phase shifted.
(3.9)
as;
cos(y)-cos5(;r)+cos7(x) - c o s l l(r^)+-^cosl3(rf)---11
13
2a_ph
(3.11)
(3.12)
VI
is:
co?{a)t - 30) - ^cos(5yr + 30) + ycos(7aY - 30)
ka-ha_ph
hc_ph~^
(3.13)
Compare equation (3.9) with (3.13), we can see that the phase displacement
5=
26
lb ) Harmonic elimination from the 12-pulse SCR rectifier primary line current.
For the 12-pulse SCR rectifier, based on Fig. 3-2(d) and equation (2.15), we can write
the Fourier series o f iasi as:
COs(^ ) - COS5 {(Dt ) + i COS 7 {(Dt) ^ COS1 \{o^) + ^ COS 13{(Dt)
(3.14)
Since ias2 lags iasi by 3 0 , thus the Fourier series o f ias2 can be written as:
cos ( >/ -
Y y c o s l
\{ (D t
V
V
Based on transform er turns ratio o f - S - =
_
^a\b\
^262
1. 1/ ,
^asl -
by using the conclusion in section la), we can write the Fourier series o f ias2 as:
cos[(^y/ - 30) + 30] - -cos[5(2/ - 3 0 ) - 30]
1.1/ ,
asl
\{ c o t
- 3 0 ) - 30]
(3.17)
^as\ +Cj2 ~ 1-l^Z, c o s (< a r)-^ c o s (l l< 5 jr)+ ^ c o s (l3 it)-^ c o s (2 3 fu t)+ ...
(3.18)
in w hich the 5* 7, 17^\ 19^' ... order harmonics are zero, and the lowest order harm onic
is the 11*.
27
For the three-phase positive sequence currents, we can write the equations as:
I = 4 sin()t),
/,= / s in
I
3 y
i; = 4 s in |I O +---3
(3.19)
(3.20)
(3.21)
For the three phase negative sequence currents, we can write the equations as:
4 = 4sin(ur),
^
ast-v
271^
^
2;r
K = L sin X -
(3.22)
(3.23)
(3.24)
For the 5^ order harmonic currents, we can write the following equations;
k5,A=4mSin5(ait),
28
(3.25)
c
h,5th h m s i l l 5 o a t -
Km = 4m sin 5 cot +
V
2 ;r l
'I t c '
= hm sm 5 6 ^ 4 3 )
V
2 ;r l
f
=
sm Scot -
3 7
(3.26)
2;r^
(3.27)
T ,
Hence the 5* order harmonic currents are negative sequence currents. Similarly, we can
prove that the (5+6^:)**^ (A==0,l,2,...) order harmonics are negative sequence currents, and
the (7+6)* (A =0,l,2,...) order harmonics are positive sequence currents.
^ ^
Ij, /j,
F ig. 3 -6
The fact that negative sequence voltages and currents are shifted in the opposite
direction to positive sequence values when passing through a phase-shifting transform er
m akes it possible to cancel harmonics in pairs.
In each line current, if the fundamental component 7, rotates at the angular frequency
CO,
the 5^ order harm onic current 7, will rotate at the angular frequency 5co, and the 7*
property
29
OF
order harmonic current /, will rotate at the angular frequency 7<y. As a result, when
rotates 10,
4 ,2
4 ,2
5,,,
4*2
is(
4^ is?
primary side, iasi has no phase shift with respect to iasi, but ias2 has different phase shift
in its reflected components:
1,,
phase shifted with respect to those in ias2 , and all negative sequence components in ias2
are -30 phase shifted. As a result, 4,2_i. becomes - 30 + 30 = 0 lagging to Ci_i5/ 1
L i _ 5 ,h is -1 5 0 -3 0 = -180 lagging to
lagging to 4 ,
and C ;
1^,2
is - 210 + 30 = -180
o f them is zero. So do 4 , 7,,, and i,2 _vh Because the line current
and 7*"^ order harmonics are zero in
except the (12t 1)"" order harmonics (A: = 1,2,3,...) . Therefore, there is only the
(12A: 1)'^ order harmonics {k = 1,2,3,...) in
the reflected harmonics being cancelled in the primary winding of the 12-pulse SCR
rectifier.
Fig. 3-7
30
ia si
_ 7-^05
+ h \\
IgU +
= 31.1%
(3.28)
a\
- ^A\ _
T H D ,,=
^A\\
+ A12/ +^^25 +
-^A\
= 15.3%
(3.29)
ax
It can be seen that the THD o f U reduces about 50% compared to that o f ias] and ias2 ,
ow ing to the elim ination o f harm onics in U-
Fig. 3-8(a) show s the harm onic content o f iasi and ias2 , where the 5* and 7*^ order
harm onics have quite high am plitudes (about 20% and 13%), and the THD o f iasi or ias2 is
28.73% . Fig. 3-8(b) shows the harmonic content o f U, in which the 5* and 7* order
harm onics are zero and the THD is 12.89%. The THD values are slightly lower than the
result presented in equations (3.28) and (3.29) due to the inductance effect.
Harmonic spectrum of
3
THD=28.73 %
d.
u
snd
0.5
jy
25th
A A.
15
10
20
25
30
Frequency (p .u .)
(a)
Harmonic spectrum of/^
T H D = 12.89 %
ci
I
0
11th 13th
A
10
23rd 25th
A,
ft
15
20
IK
25
30
Frequency (p .u .)
(b)
Fig. 3-8
31
Fig. 3-9 shows the THD of ia and the input PF of the 12-pulse SCR rectifier versus Ui
under different operating conditions. Compared to the plot of the six-pulse SCR rectifier
shown in Fig. 2-7, the THD of the 12-pulse rectifier is reduced and the input PF is
increased. However, the THD values still do not meet the harmonic current requirement
of IEEE Standard 519-1992.
T H D o f I , at a = 0
T H D o f t , at L = 0 .0 5 p .u .
15
u. 1
g,4
0.8
13
12
1^=0 05 p.u.
0
0.2
0.4
0.6
Fig. 3-9
0.8
1
/^,(p.u.)
11
A
B
'
C
C. a=60
B:a=40
A: a=20
L^=0.05 p.u.
0
0.2
0.4
0.6
N.
^
0.6
0.8 1
/..(p.u.)
A:a=0
B: 0=20
0.4 C:a=40
Lg.=0.05p.u.
0
0.2 0.4
0.6 0.8
(4; (pu)
THD and PF curves o f the 12-pulse SCR rectifier with inductive loads.
5 = 0
5 = -30
Fig. 3-10
32
i.e.
CLaaJ
(b) transformer secondary side currents
Fig. 3-11
33
Fig. 3-12 shows the waveform and harmonic spectrum of Iai = 1 pu (rated load
condition) and Iai = 0.2 pu (light load condition), respectively. It can be seen that the Ia
waveforms are quite different since the energy exchanged through Ls varies when the
load current changes. The 5^', 7^*, 17*' and 19*' order harmonics are zero in Ia owing to
the phase-shifting transformer.
p.u.
THD=11.89%
1.414
0.5
11th
JL Z L
-40
10
(a)
15
23rd 25th
**
20
25
30
Frequency (p.u.)
2^,=0.2 p.u.
THD=24.37%
O.
0.282
0
0.5
A.
-10
(b)
Fig. 3-12
11th
23rd 25th
10
15
20
25
30
Frequency (p.u.)
34
In Fig. 3-13(c), a crossover occurs at approximately I ai = 0.22 pu. Its the boundary
o f the continuous and discontinuous current operation mode. W hen Iai is greater than
0.22 pu, the load current will be operating in the continuous mode.
TH D o f /, at Lj=0.05p.u.
TH D o f / , at a= 0
g , 25
P 20
40
Q
A: Lj=0 p.u.
B: Lj=0.05 p.u.
C :L j= 0.1p .u .
Ln=0.05 p.u.
\
\
\
\
30
V\
\ \
\ \
\
\ \
u,
a.
C :a = 6 0
B: a = 40
A: a=20"
Lg.=0.05 p.u.
0.98
15
20
A: Lj=0 p.u.
B: L =0.05 p.u.
C: L = 0.1 p.u.
Lg.=0.05 p.u.
0.96
10
5
10
0
0.2
0.4
0.6
0.8
0.2
0.4
Fig, 3-13
0.8
0.94
02
0.4
0.6
(p.u.)
I ., (p.u.)
(a)
0.6
(b)
'
(c)
35
0.8
1
/^ ,(p .u .)
3.3 Summary
This chapter presents the performance of the 12-pulse SCR rectifier with inductive
and capacitive loads. The following eonclusions can be made:
The principle of the harmonic elimination through the phase-shifting transformer can
be illustrated by Fourier analysis and the positive/negative sequence analysis.
For the inductive load condition, the ripple in the dc load voltage has a doubled
frequency and reduced peak-to-peak amplitude compared to the six-pulse rectifier.
The RMS value of the line current is lower than that of the six-pulse rectifier due to
the harmonic elimination.
For both of the load conditions, the line current THD values are lower than those of
the six-pulse rectifier, because the dominant low-order harmonics are eliminated.
The input PF of the 12-pulse rectifier is improved.
The line eurrent THD of the 12-pulse SCR rectifier with either the inductive or the
capacitive load still cannot meet the harmonic current requirement in IEEE Standard
519-1992.
36
Since the line current THD o f the 12-pulse SCR rectifier does not m eet the harm onic
current requirem ent in IEEE Standard 519-1992, it is required to elim inate more loworder harm onic currents. The 18-pulse SCR rectifier is an option to im prove the line
current THD.
This chapter presents the perform ance o f the 18-pulse SCR rectifier w ith inductive
and capacitive loads. The voltage and current waveform s are shown. The harm onic
spectrum o f the line current and the reason o f harmonic elim ination are presented. The
line current THD and the input PF are discussed.
37
5 = -20
5 = -40
Fig. 4-1
th
11*' and 13" order harmonics are eliminated; thus the lowest order harmonic is the 17
and the THD is 7.45%.
In Fig. 4-2(d), vn, v u and v/j are output voltages of the three six-pulse SCR bridges
respectively. Their waveforms are the same as that of the typical six-pulse bridge, but are
20 phase shifted. Their summation is the load voltage
vl,
is tripled compared to Vu, Vu and vu. Hence the 18-pulse rectifier requires a smaller dc
choke to obtain a ripple free dc load current compared to the 12-pulse rectifier.
38
-40
(a) Transform er primary sid e current/^
15
= L /-)
r asS
'a s /
'" i r z F ^
i
-15
(b)
40
N-W-H
20 20
-40
(c)
and
300
\A(NNA/NNNV)A/NNNV^
3
d.
a=30
THD=7.45%
U 0.5
17th 19th
A Aw
10
15
20
25
30
Frequency (p .u .)
F ig . 4-2
39
120
> 0 windings 5"', 7"', ll"'and 13"'
-
120
Fig. 4-3
40
& '
A
B
H 9
0.8
C
0.6
6
0.2
0.4
0.6
0.8
(4; (p.u )
F ig. 4 -4
A: a=0
B: a=20
0.4 C: a=40
L,^=0.05 p.u.
C: a=60
B: a=40
A: a=20
Lj|.=0.05 p.u.
0.2
0.4
0.6
0.8
1
/^/p.u.)
0.2
0.4
0.6
0.8
a s /
~~as2
'^ 'a s J
0 = 0
5 = -20
5 = -40
Fig. 4-5
41
/^,(p.u.)
and as3 are the result of the zigzag connections. The six small ripples on the positive half
cycle o f each secondary line current are shown in Fig. 4-6(c). It's the result of the
involvement of the three transformer secondary windings and the three six-pulse SCR
bridges.
-40
(a) Transformer primary side currently
15
f
a s2
^ as3
XXA
-15
(b) Transformer primary side currents;
; .'and i
'
40
'as2
0
20 20
-40
(c) Transformer secondary side currents
and the load current i.
'j
a=30
THD=7.92%
0.5
17th 19th
A &
5
(d)
Fig. 4-6
10
15
20
25
30
Frequency (p.u.)
Harmonic spectrum o f
42
w hen a=20, I aj = 1
pu (rated load condition) and I ai 0.2 pu (light load condition). It can be seen that the 5*,
7*^, 11* and 13* order harm onics are eliminated, and the am plitude o f the current ripple is
substantially reduced, com pared to the 12-pulse SCR rectifier. The THD values are
6.036% (/^/=1 pu) and 12.5% (Li/=0.2 pu), which are less than those o f the 12-pulse
rectifier under the same conditions.
PU-
THD=6.036 %
'A !
1.414
0.5
17th 19th
* *-
-40
5
(a)
1
15
20
25
30
Frequency (p.u.)
'A !
10
0.282
0.5
17th
JL_
-10
(b)
F ig. 4-7
10
15
20
25
30
Frequency (p.u.)
Fig. 4-8(a) and Fig. 4-8(b) show the Ia THD at different value o f L s ,
Ia i
and a. U nder
the condition o f a = 0 and 7^/>0.5pu, the THD values are less than 5%. Fig. 4-8(c) shows
the input PF o f the 18-pulse SCR rectifier with capacitive loads. The values are im proved
slightly com pared to that o f the 12-pulse rectifier. Since three transform er secondary
w indings are connected in series, the total transform er leakage inductance becom es larger.
Therefore the load current operates in the continuous mode.
43
S'
CIO-------------^
5 20
Q
g
\ \
15
10
0.2
0.4
0.6
0.8
0.98
A:Lj=Op.u.
0.96 B: Lj=0.05 p.u.
C: Lj=O.I p.u.
Lg.=0.05 p.u.
0.94
0.2
(a)
Fig. 4-8
V'
C:a=60
B:a=40
A; a=20
L,=0.05 p.u.
0.4
0.6
0.8
1
/^,(p.u.)
0.2
0.4
0.6
0.8
(b)
(c)
THD and PF curves o f the 18-pulse SCR rectifier with capacitive loads.
4.3 Summary
This chapter presents the performance of the 18-pulse SCR rectifier with the
inductive and capacitive loads. The following conclusions can be made:
The principle of the harmonic elimination can be applied to the 18-pulse rectifier.
For the inductive load condition, the ripple amplitude in the dc load voltage is
reduced, and the ripple frequency is tripled compared to that of the six-pulse rectifier.
For the capacitive load condition, since the leakage inductance of the three
transformer secondary windings is in-series, the load current operates in the
continuous mode.
For the two load conditions, the THD values are substantially reduced, because more
low-order harmonics are eliminated.
For the two load conditions, when the delay angle a=0 and
1a i > 0 . 5
current THD is less than 5%, which meets the harmonic current requirement of IEEE
Standard 519-1992.
44
Iai (P )
In order to further reduce the line current THD and handle higher voltages, 24-pulse
SCR rectifiers are employed. This chapter presents the perform ance o f the 24-pulse SCR
rectifier w ith inductive and capacitive loads. The line current THD and the input PF o f
the 24-pulse rectifier are discussed.
45
II.,
X ^asi
yo-V
5 = 0
+
V,o
5 = -15
-30
(^^^L-rY-y~>rK
5 = -45
Fig. 5-1
' +w ' +
be seen that the 5*'^, 7*'', 11*', 13", 17*' and 19*' order harmonics are zero in h; thus the
lowest order harmonic becomes the 23"^". As shown in Fig. 5-2(a), the waveform of a is
almost a sinusoid with 5.09% THD. In Fig. 5-2(d), the pulsated dc load voltage vl has a
higher ripple frequency and a lower ripple amplitude, compared to that of the 18-pulse
rectifier; therefore, a ripple free dc current can be obtained easier from the 24-pulse
rectifier.
46
-4 0
(a) Transformer primary side current;^
10
^ as3
- iql
(b )
'
40
N-NHR
151515
-40
(c)
?
a
'
a=3QP
T H D = 5.09%
o 0.5
23rd 25th
10
(e)
F ig. 5-2
15
20
25
30
Frequency (p .u .)
47
Fig. 5-3
48
THD ofi^ at a -0
&
A: Lj=Op.u.
&
B; Lj=0.05 p.u.
C: Lj=0.1 p.u.
L^=0.05 p.u.
0.2
0.4
0.6
0.8
I 7
0 .8
5 C: a=60
B: a=40
4 A: a=20
L^=0.05 p.u.
3
0 0.2 0.4
0 .6
^S*A
0.6
^A! (P " )
Fig. 5-4
5.2
0.8
1 .,
A: a=0
B: a=20
C: a=40
L|^=0.05 p.u.
0.4
0 -2
0.2
0.4
0.6
(p.u)
0.8
Iai (P
e x
5 = 0
5 = -15
S = -3 0
Fig. 5-5
5.2.2 C u r r e n t W av efo rm s
Fig. 5-6 shows the current w aveform s under the condition o f a=30,
Ia i
= lp u ,
Ls
=0
and Lik =0.05pu. The line current Ia in Fig. 5-6(a) is alm ost a sinusoid w ith little ripples.
Fig. 5-6(b) shows the w aveform s o f iasi\ ias2 \ iass and ias/, hi w hich ias2 and ias4 are
caused by zigzag connections. In Fig. 5-6(c), it can he seen that the ripple am ount in the
49
'>-)
positive half cycle of each secondary line current is increased to eight and the ripple
amplitude is further decreased. Hence a ripple free dc load current can be obtained easier.
The harmonic spectrum of a is shown in Fig. 5-6(e), in which the lowest order harmonic
is the 23'^^, and the THD is 5.2%.
-40
(a) Transformer primary side current
'as2.
-10
(b) Transformer primary side currents
^as2
^as2
os7
as4
-40
(c) Transformer secondary side currents
and load current i,
3.
a
a=30
THD=5.2 %
U0.5
!L
23rd 25th
10
15
(d)
Fig. 5-6
20
25
30
Frequaicy (p.u.)
Harmomc sp ectrum of
50
th^
17* and 19* order harm onics are elim inated, the TH D o f Ia is 3.62% and
7.55% at Iai 1 pu and Iai = 0.2 pu, respectively, which are lower than those o f the 18pulse rectifier.
Fig. 5-8 shows the line current THD and the input PF o f the 24-pulse SCR rectifier
w ith capacitive loads under different operating conditions. It can be seen that the THD
values are further reduced. Especially when a=0, the TH D values satisfy the IEEE
Standard 519-1992. Under heavy load conditions, the PF o f the 24-pulse rectifier is
slightly reduced com pared to that o f the 12- or 18-pulse rectifiers, since the phase
displacem ent o f the line current caused by the line inductance and the transform er
leakage inductance is slightly greater.
Harmonic spectrum o f
30
7^,=1 p.u.
T H D = 3 .6 2 %
1.414
0.5
23rd 25th
-30
10
15
20
(a)
Harmonic spectrum o f
25
30
Frequenty (p.u.)
1^1 = 0 2 p.u.
T H D =7.55%
'A !
0.282
0.5
23rd 25th
- A - __
-10
(b)
Fig. 5-7
10
15
20
25
30
Frequency (p.u.)
51
THDof)^ata=0
C:a=60
B; a=40
A: 0=20
L1^=0 05 p.u.
\\
\ \
A;L,
B:L.
C: Lj=0.1 p.u.
L||^=0.05 p.u.
0
0.2
0.4
0.6
0.8
0.2
0.4
(4/ (Pu )
Fig. 5-8
0.6
0.8
1
(4 , (p.u.)
Input PF at 0=0
'
0.98
A; Lj=0 p.u.
0.96 B: Lj=0.05 p.u.
C: Lj=0.1 p.u.
L||,=0.05 p.u.
0.94
0 0.2 0.4
0.6
0.8
THD and PF curves o f the 24-pulse SCR rectifier with capacitive loads.
5.3 Summary
This chapter presents the performance of the 24-pulse SCR rectifier with inductive
and capacitive loads. The following conclusions can be made:
The principle of the harmonic elimination can be applied to the 24-pulse rectifier.
For the two load conditions, since more low-order harmonics in the line current are
eliminated, the line current THD of the 24-pulse rectifier is further reduced.
Especially when the delay angle a=0, the THD values meet the harmonic current
requirement of IEEE Standard 519-1992.
For the capacitive load under heavy load conditions, the PF of the 24-pulse rectifier
is slightly reduced, because the phase displacement of the line current caused by the
line inductance and the transformer leakage inductance is slightly greater than that of
the 12- or 18-pulse rectifier.
Multi-pulse SCR rectifiers with a higher pulse number, such as the 30- or 36-pulse,
are rarely used in industry because of the high cost.
52
m (p.u.)
In previous chapters, the perform ance o f m ulti-pulse SCR rectifiers was presented
and the principle o f the harm onic elim ination through the phase-shifting transform er was
discussed. This chapter presents the experim ental verification on a prototype o f the 12pulse S C R rectifier with an inductive load. The related hardware design and software
program m ing are provided. Typical voltage and current w aveform s obtained from the
experim ent are com pared to the sim ulation results.
3-phase ac
p o w er supply
(208V )
12-pulse SCR
rectifier
Load
Step-dow n
transform er
(120V /6V )
- 3
Zero-crossing
detection board
12V
PC
Firing
board
68HC11 based
platform
Power circuit
5V
dc power
supplies
Control circuit
C ontrol C ircuit
Fig. 6-1
53
Fig. 6-1 shows the hardware configuration of the experimental 12-pulse SCR rectifier,
which consists of the power circuit and the control circuit.
Power Circuit
The power circuit o f the 12-pulse SCR rectifier includes a three-phase sinusoidal ac
power supply, an inductor bank Ls, a phase-shifting transformer (A | AY), two six-pulse
SCR bridges, and a load.
The inductor bank Ls'. Since the internal impedance of the three-phase sinusoidal
power supply is low, the inductor Ls is needed for emulating the actual occasions.
SCR rectifier. The 12-pulse SCR rectifier has two six-pulse SCR bridges, whose dc
outputs are cormected in series.
54
Load: The load o f the 12-pulse rectifier is a resistor bank connected in series w ith a
4 0 m H mdvLcXor bank.
A
B
C
c2
Fig. 6-2
Control Circuit
The control circuit consists o f a step-down transform er (120V/6V), a zero-crossing
detection board, a M C 68H C 11 control board, a SC R firing board, dc pow er supplies and
a personal computer.
M C 68H C 11 control board: The SCR gating pulses o f the 12-pulse rectifier are
generated by the M C 68H C 11 control board.
F iring board: The gating pulses generated by the M C 68H C 11 control board are
am plified and isolated by the firing board to m eet the triggering requirem ent o f SCRs.
D c p o w er supplies: the dc pow er supplies provide 5V, 12V and 15V dc voltages to
the control circuit.
55
Personal computer, the personal computer instructs the MC68HC11 control board to
realize the real-time control.
average output voltage of each six-pulse SCR rectifier under ideal circuit condition is
^Li(2 ),owg - 1 -35f}y cosa = 1.35 X 120 X cosa = 162cosa, where a is the delay angle. From
Equation (3.8), the average output voltage of the 12-pulse rectifier with an inductive load
is
=1.93xl62cosr = 313cos
=271/40 = 6.78/1
and
the
average
= 4 0 0 and a - 30,
Considering
the
load
current
is
= 313cos30 = 271E
transformer
turns
ratio,
we
V3
208
= j=^ = 120V ;
V3
P'aT 120
the base impedance Z t = ^ = --------= 2.160;
5 5J6
the base angular frequency 0)gj = 2 ?rf = 2 7 1 x 6 0 = 7 ,1 1 rad / j ;
56
2^
2 1fi
the base inductance L r = ^ = =
(Og-,
377
l^ m H
= --= 1 .2 2 S m F .
3 7 7 x 2 .1 6
real value
= 0.235m // , or
o f the total
= 4 . 1 % x 5 .7 3 w / /
= 0.235m H
= 2.2K V A
= 69.28F ;
v3
c
V3
2700
= 17.3 6 m // ;
377
17.36m //
= 0.0135pw .
Ls Calculation
A three-phase inductor bank is added to the circuit to em ulate the actual occasions.
Choose Z, = Im H ; then based on the experim ental system, Z^ = Im H =
= 0.057O p u , w hich is a reasonable value.
57
17.36/w/z
Ia THD Calculation
From Equation (2.18), THD^^ _ V
a ,r m s
\ , ^
the
^A \,R M S
setup can be calculated based on Ia.rmsjbxp, the RMS value of the line current, and
hi,RMS_exp, the RMS value of the fundamental component of the line current. By using the
measurement function of the oscilloscope, we have h.RMSj^xp- Through the FFT function,
we can get the harmonic spectrum of the line current, of which the measured peak value
is lAi,ms_exp- Thus the Ia THD of the experimental system can be obtained.
58
C h l R M S 1 1 8 V C h2 R M S 1 1 6 V
Fig. 6-3
Fig. 6-3 shows the line-to-line voltage waveform s o f the tw o transform er secondary
w indings, in w hich a 30 phase-shift can be observed. The m easured RM S line-to-line
voltages o f the two windings are 118V and 116 V, respectively.
Z e ro -C ro ssin g S ignal
In Fig. 6-4, the sinusoidal waveform is the line-to-line voltage Vac o f the wyeconnected secondary winding, and the square waveform is the output signal vzc o f the
zero-crossing detection board. Vac is the phase A zero-crossing reference signal, whose
zero-crossing points are corresponding to the nature com m utation points o f phase
in a
three-phase system. H ence the delay angle a could be m easured w ith respect to the v,ac
zero crossing points, w hich are corresponding to the rising and falling edges o f vzc-
59
Fig. 6-4
"zc -
glj
glj
S3J
g3J
i l j -
t 6 J
(a)
Fig. 6-5
(b)
60
It can be seen that the tw o sets o f gating pulses are phase-shifted by 30, w hich m eets
the gating requirem ent o f the 12-pulse SCR rectifier. For the first gating pulse igi u the
delay angle a is measured right after the rising edge o f vzc- The phase displacem ent
betw een any adjacent gating pulses in each set is 60.
17** and 19*** order harm onics exist in the two currents and
produce a TH D o f 34.4%.
Fig. 6-6(c) shows the m easured current waveform o f the transform er prim ary
w inding, w hich looks more like a sinusoid com pared to Fig. 6-6(a) and Fig. 6-6(b). From
the harm onic spectrum , it can be seen that the 5*, 7**, 17** and 19** order harm onics are
elim inated. Therefore the THD o f iz is reduced to 17.52%.
Fig. 6-6(d) shows the m easured output voltage w aveform s o f the two six-pulse SCR
bridges. It can be m easured that the phase shift between the two voltages is 30. Fig. 66(e) show s the m easured dc load voltage waveform o f the 12-pulse SCR rectifier.
C om pared to those in Fig. 6-6(d), the dc ripple frequency in Fig. 6-6(e) is doubled. Fig. 66(f) show s the m easured dc load current, which has a m ean value o f 5.41 A.
Since losses and fluctuations exist in the real system, the experim ental results in Fig.
6-6 have a low er line current RM S, a higher line current TH D, and a lower m ean dc load
current th an the sim ulation results.
61
6.5 Summary
This chapter presents the experimental verification of the performance of multi-pulse
SCR rectifiers on a prototype of the 12-pulse SCR rectifier with an inductive load. From
the experimental results, we can get the following conclusions:
The 5*',
17^' and 19*' order harmonics exist in the transformer secondary winding
line currents.
The 5*'',
17*' and 19*' order harmonics do not exist in the transformer primary
The line current THD of the 12-pulse SCR rectifier is lower than that of the six-pulse
SCR rectifier.
The dc ripple frequency of the 12-pulse rectifier is doubled, compared to that of the
six-pulse rectifier.
62
Simulation Result
r j- 4 .8 9 A
-10
1
THD = 29.07 %
0.5
C h l 5.00A
C h lR M S 4 .7 8 A
M a th 2 .0 0 A 1 2 5 H z
M a th M a x 4 .5 2 A T H D = 3 4 .4 %
ISt ,
-10
1
THD = 29.09 %
0.5
10
C h l 5. OOA
C h l R M S 4 .78A
15
20
M a th 2. OOA 1 2 5 H z
M a th M a x 4 .5 2 A T H D = 3 4 .4 %
-10
-20
1
THD - 13.3%
0.5
M a th 2. OOA 12 5 H z
C h l 5.0 0 A
C h l R M S 9.33A M a th M a x 9 .1 9 A T H D = 1 7.52%
10
15
20
Simulation Result
Experimental Result
200
100
300
250
200
=5.946A
Chn.OOA
ChlMean 5.4M
Chapter 7 Conclusions
M ulti-pulse SCR rectifiers are widely used in high power applications in industry due
to their low line current Total Harm onic Distortion (THD) and high voltage capabilities.
The m ain contribution o f the thesis is to analyze the performance o f the six-, 12-, 18- and
24-pulse SCR rectifiers with inductive and capacitive loads. The line current THD and
the input Pow er Factor (PF) o f the rectifiers are investigated and the results are shown in
graphics for engineering references. The principle o f the harmonic elim ination through
the phase-shifting transform is analyzed by Fourier analysis and positive/negative
sequence analysis. Experim ental verification is completed on a prototype o f the 12-pulse
SCR rectifier.
U nder the same operating condition, the larger the num ber o f pulses, the more the loworder harm onics can be eliminated, and the lower the line current THD is. For instance,
the TH D o f the 18-pulse SCR rectifier is lower than that o f the 12-pulse rectifier.
The output dc ripple frequency in the multi-pulse rectifiers is increased and the ripple
am plitude is decreased.
The dc ripple frequency o f the 12-, 18- and 24-pulse rectifiers is increased to two, three
and four tim es o f that in the six-pulse rectifier, respectively. The peak-to-peak dc ripple
am plitude decreases as the pulse num ber o f the rectifier increases. Therefore, it is easier
to obtain a ripple free dc output in the rectifier with a larger pulse num ber.
65
When SCRs are connected in series to handle high voltages, simultaneous operating of
the SCRs has to be ensured. In multi-pulse rectifiers, the SCR bridges are connected in
series rather than SCRs, and hence the voltage sharing problem of the in-series SCRs is
avoided. The larger the number of pulses, the more SCR bridges should be connected in
series, and the higher voltage the rectifier can handle.
where q is the
In order to verify the performance of the multi-pulse SCR rectifiers, a prototype of the
12-pulse SCR rectifier controlled by a microprocessor is designed and constructed. The
experimental results show that the low-order harmonics, such as the
eliminated firom the line current of the 12-pulse SCR rectifier. As a result, its line current
THD is reduced substantially compared to that of the six-pulse rectifier.
66
TOC3 E Q U $101A
T O C 4E Q U $101C
TOC5 EQU $ 1 0 IE
TM SK l EQU $1022
DD R D EQU $1009
;port A
;portD
ORG $6000
67
89
00001100% 9 3 J
00000100% 9 3 d
00000100% 9 3 d
00000000% 9 3 d
00000000% 9 3 d
00000000% 9 3 d
00000000% 9 3 d
00010000% 9 3 d
00010000% 9 3 d
00011000% 9 3 d aO NIlVO
00000000% 9 3 d
00000000% 9 3 d
00000000% 9 3 d
00001000% 9 3 d
00001000% 9 3 d
00001100% 9 3 d
00001100% 9 3 d
00000110% 9 3 d
00000110% 9 3 d
00000010% 9 3 d
00000010% 9 3 d
00000000% 9 3 d VDNllVO
01100000% 9 3 d
01100000% 9 3 d
00110000% 9 3 d
00110000% 9 3 d
00011000% 9 3 d
00011000% 9 3 d
00001100% 9 3 d
00001100% 9 3 d
00000110% 9 3 d
00000110% 9 3 d
01000010% 9 3 d
01000010% 9 3 d 99NL1V0
F C B % 00110000
FCB % 00011000
PATTI FCBO
NOCl
FCBO
;Iine feed
FCB $04
;string terminator
C R T M SG l
FCB $0A
FCB $04
L C D M SG l
ENDM SG
FCB $04
;LCD message
BUFFER
RMB 3
ALPH A D FCB 10
D 30 EQU 2778
ORG $00E8
JM PINTER6
ORG $00E5
JMP INTER7
69
ORG $00E2
JMP INTERS
ORGSOODF
JMP INTER!
ORG $OODC
JMP INTER2
ORG $00D9
JMPINTER3
ORG S00D6
JMP INTER4
ORG $00D3
JMP INTERS
ORG $6500
LDS #$7FFF
LDAA #$80
STAA PACTL
LDAA #$38
STAA DDRD
LDAA #$89
STAA TFLGl
LDAA #$01
STAA TCTL2
LDAA #$89
STAA TMSKl
LDAB #30
LDAA #92
MUL
D=AxB
STD#D30
70
CLI
M AIN
;interrupt enable
L D X #C R T M SG l
L D A A 0,X
JSR OUTSTR
JMP MAIN
GETCHAR LDAB #0
STAB N U M l
L D Y #KB_ALPHA
CM PA #$39
BHI ERROR
CM PA #$30
BLO ERROR
STA A 0,Y
LDAB N U M l
INCB
CMPB #4
71
BHS ERROR
STAB NUMl
INY
BRA START
CHECKl
LDAA NUMl
CMPA#1
BEQ CHECK2
CMPA #2
BEQ ENDD
LDAB KB_ALPHA
CMPB #$31
BEQ CHECKS
BRA ERROR
CHECK2
LDAA KB_ALPHA
CMPA #$31
BLO ERROR
JMP ENDD
CHECKS
LDAAKBALPHA+l
CMPA #$37
BLO ENDD
BHI ERROR
LDAA KB_ALPHA+2
CMPA #$39
BHI ERROR
JMPENDD
ERROR LDX #CRTMSG2
LDAAO,X
JSR OUTSTR
BRA GETCHAR
ENDD RTS
LDAA #$01
;LCD initialization
JSR WCTRL
LDAA #$02
72
JSR WCTRL
L DA A #$38
JSR WCTRL
LD A A #$0C
JSR WCTRL
LDY #LCDMSG1
LOOP2 CPY #ENDM SG ;compare index register Y,
BEQ NEXT5
LDAA 0,Y
JSR W DAT
INY
L DA A #$06
JSR WCTRL
B R A LOOP2
L D Y #KB_ALPHA
LO OPl L D A A 0,Y
JSR W DAT
INY
INC BUFFER
L D A A BUFFER
CM PA N U M l
BEQ NEXT6
LD A A #$06
JSR WCTRL
B R A LOOPl
NEXT6 L D A A #'D'
;display 'D'
JSR W DAT
RTS
73
CMPA #1
BEQ SINGLE
CMPA #2
BEQ DOUBLE
LDAA KB_ALPHA
SUBA #$30
LDAB #100
MUL
STAB BUFFER
LDAA KB_ALPHA+1
SUBA #$30
LDAB #10
MUL
ADDB BUFFER
STAB BUFFER
LDAA KB ALPHA+2
SUBA #$30
ADDA BUFFER
STAA ALPHA_D
;ALPHA_D=lxx degree
BRA CALULAT
SfNGLELDAA KB ALPHA
SUBA #$30
STAA ALPHA_D
;ALPHA_D=x degree
BRA CALULAT
DOUBLE
LDAA KB_ALPHA
SUBA #$30
LDAB #10
MUL
STAB BUFFER
LDAA KB ALPHA+l
SUBA #$30
ADDA BUFFER
STAA ALPHA D
CALULAT
;ALPHA_D=xx degree
LDAA #92
LDAB ALPHA D
74
MUL
;D=AxB
STD ALPHA
RTS
;bridgel
LD A A #$01
;cIearIC3F
STAA TFLGl
RTI
LDDTOC5
A D D D #D30
STD TOCl
LDAB #$08
STAB TFLGl
RTI
75
LDAB PATTI
INCB
LDY#GATINGB
ABY
LDAA 0,Y
STAA $1100
LDY#GATINGA
ABY
LDAAO,Y
STAA PORTA
LDY#GATINGD
ABY
LDAAO,Y
STAA PORTD
CMPB #10
BLS NEXTl
LDAB#0
NEXTl STAB PATTI
LDD TOCl
ADDD #D30
STD TOCl
BRA QUIT
LASTPUL
LDAB #0
STAB NOCl
QUIT
LDAB #$80
STAB TFLGl
RTI
76
50
a=<r
T H D o f /^ a t L j= 0 .1 p ii.
^
51 :
------
------------------------------------------------------
u.
1
A
Q
28
0 8
29
26
C
0 6
28
24
B
A : L j,= 0 .0 5 p .u .
B : L j = 0 .1 0 p .u .
C : L = 0 .1 5 p .u .
22
i
27 :
>
0 .4
0 .6
0.8
(P U)
A : = 0
B: = 20
C: =40
0 .4
A: =20
2 6 ----------------------------------- ------------ '----------0
0 .2
0 .4
0 .6
0 .8
1
20
0 .2
V .
C: =60
B: =40
0 .2
0
0 .2
0 .4
0 .6
0 .8
/.,(P u)
(P u )
T H D o f /^ a t L ^ = 0 .0 5 p .u .
T H D o f / , at a = 0
A
B
0.8
C
0.6
C: a= 6 0
B: = 4 0
A: =20
L |^ = 0 0 5 p .u
A : L j = 0 p .u .
B : L j = 0 .0 5 p
C : L ^ = 0 .1 p . u
L ||.= 0 .0 5 p .u
A: =0
B: = 20
C: =40
L |i,= 0 .0 5 p . u .
0 .4
0.2
0 .2
0 .4
0 .6
( p .u .)
0 .8
/m;CP u.)
1 2 -p u k e w ith inductive lo a d s
I n p u t P F a t L ,= 0 .0 5 p . u .
T H D o f ; . a t L j = 0 .0 5 p .u .
10
Q
A : L j = 0 p .u .
B : L j = 0 .0 5 p .u
fc
C ; L j= 0 .1 p .u .
L , = 0 .0 5 p .u
'
0 .8
8
0 .6
C: =60
B: = 40
A: =20
L ,|.= 0 .0 5 p .u.
7
6
0 .2
0 .4
0 .4
'
0 .6
A: =0
B: = 2 0
C: =40
L ,|^ = 0 .0 5 p . u .
0 .8
I
/,,( p .u .)
0 .2
0 .4
0 .6
0 .8
I A, ( P )
T H D o f /, at = 0
&
A :L j= O p .u .
B : L ^ = 0 .0 5 p .u .
X
4
I n p u t P F a t L ,= 0 .0 5 p .u .
T H D o f /,, a t L j = 0 .0 5 p . u .
A
B
'
a
S
L ,.= O .O S p .u .
"
^^ * * * ' ^ __
0 .8
6
0 .6
5
C ; = 6 0
B: =40
A : =20
L ,^ = 0 .0 5 p . u .
2
4
1
0 4
0 2
0 .2
0 .4
0 .6
0 .8
0 .2
0 .4
0 .6
0.8
/ . . ( p .u .)
(t,( p " )
2 4 - p u b e w ith in d u c tiv e lo a d s
77
A: =0
B: = 20
C: =40
L ^= 0 .0 5 p . u .
0 .2
0 ,4
0 .6
0 .8
' a , <P u .)
THDof/at 0=0
A: Lj=0.05p.u.
B: L,=0.10p.u.
C: L=0.15 p.u.
THD cfi;atL,=0.1p.u.
S' 140
C:a=60"
B; a=40
A- a
20
0.9
0.8
0.7
0.2 0.4
0.6
0.8
L
I
(P " )
0.2
0.4
0.6
0.8
1
L i (P " )
0.6
&25
g
F 20
: 40
Q
g
30
0.98
15
A; Lj=0 p.u.
0.96 B: Lj=0.05 p.u.
C; Lj=0.1 p.u.
L,^=0.05 p.u.
0.94
20
10
5
0
0.2
0.4
10
0.6
0.8
0.2 0.4
0.6
0.8
/..(p.u.)
/x,(Pu)
(./(pu)
THDof/^ata=0
>10
C: a=60
B:a=40
A: a=20
Ln=0.05 p.u.
\ Wc
15
Input PFata=0
0.98
10
0.2
0.4
0.6 0.8
/.4,(P u)
0.2
0.4
0.6
0.8
Li (Pu)
L;=Op.u.
0.96 A:
B: Lj=0.05 p.u.
C: Lj=0.1 p.u.
L|i.=0.05 p.u.
0.94
0.2 0.4 0.6
0.8
(i;(P)
^ 12 ----- VA------\\
Q
\ \
S
\ \
9
\s S ^ A
A:L,=Op.u.
B:L =0.05 p.u.
C:Lj=0.1p.u.
Lg.=0.05 p.u.
0
0.2
0.4
C: a=60
B: a=40
A: 0=20
L||.=0.05 p.u.
0.98
0.96 A:Lj=0p.u.
B;Lj=0.05p.u.
C: Lj=0.1 p.u.
Ln_=0.05 p.u.
0.94
0 0.2 0.4 0.6
0.6 0.8
tu
C
L. 1
(</(P)
L i (P U)
24-pulse with capacitive b a d s
78
0.8
1
/^,(p.u.)
References
[1] IEEE recom m ended practices and requirem ents for harmonic control in
electrical pow er systems, IEEE Standard 519-1992, pp 78,22,53.
[2]
79
80