Massachusetts Institute of Technology

Department of Electrical Engineering and Computer Science

v=1
i
eaf = 2.721
Figure1: VoltageVectorDiagramofparta
Tostart, the power factorangle is=cos
1
(0.8) 36.87

. Wecanusethelawofcosinesto
ndrequiredinternalvoltage:
e
af
2
=v
2
+(xi)
2
+2vxisin
or,equivalently,wecouldusethepythagoreantheorem:
e
2
= (v+xisin )
2
+(xicos )
2
af
whicharetransparentlyidentical.
Torqueangleis,prettydirectly:
1

xicos
= tan
1
v+xisin
Theseevaluateto:
e
af0
= 1 +2
2
+2 2 0.6 2.721
= tan
1
.6
36.03

1 +1.2
IfI
fnl
= 1000A,thenrequiredeldcurrentforthatoperatingpointwouldbe2721A.
Part b:
Sincep=
ve
af
sin,ifwevaryeldcurrentatconstantload:
x
d
= sin
1
px
d
ve
af
Notethatweareusingthesymbolp toindicateper-unitrealpower. Thestabilitylimitis
reachedwhenpx
d
=ve
af
,ore
af
= 1.6per-unit. ThatmakesI
f
= 1,600A. Anglevs. eld
currentisshowninFigure2
Problem 9.5, part b
90
1.6 1.8 2 2.2 2.4 2.6 2.8 3
80
70
60
50
40
30
PerUnit E f
a
Figure2: Problem9.5,Partb
Parts c-e Therestofthisproblemiscarriedoutinthescriptthatisappended. First,the
vee curve from the point we just computedto the overexcitedeld limit is shown in Figure 3.
Thezero-power vee curve has zero real power, so all that is interesting is reactive power:
q =
ve
af
v
, and for zero real power the absolute value of this is also per-unit armature
x
d
current (ifv= 1).
2
D
e
g
r
e
e
s

|
I

|

a

x 10
4
Problem 9.5, part b
2.5
2.4
2.3
2.2
2.1
2
1.9
1.6 1.8 2 2.2 2.4 2.6 2.8 3
PerUnit E f
a
Figure3: Problem9.5,partc
Findingtherestofthecurvesisstraightforward: thebeginningofthecurveisatanangle
determined bythe real powerand bythe maximumexcitation. The endofthe curve isat the
stabilitylimit,whichforaroundrotormachineisatanangleof90degrees.

min
= sin
1
px
d
ve
af0

max
=
2
Then we set up a vector of closely spaced points in angle, nd required value of internal
voltagetomaketherightamountofrealpowerandthencomputereactivepower:
px
d
e
af
=
vsin
ve
af
p = sin
x
d
2
ve
af
v
q =
x
d
cos x
d
i
a
= p
2
+q
2
This calculationcan be vectorized in MATLAB,andthe set ofvee curves that result is shown
ingure4.
Finally, the capabilitycurve can simply be sketched. The vector shown is drawn to the rating
point, or p=.8,q=.6.The other limits are justsections ofcircles. This is shown in Figure 5
Hereisthescript:
3
2
A
r
m
a
t
u
r
e

C
u
r
r
e
n
t

M
a
g
n
i
t
u
d
e
,

A

R
M
S

x 10
4
Problem 9.5, part e
2.5
1.5
1
0.5
0
Field Current, A
0 500 1000 1500 2000 2500 3000
Figure4: VeeCurves: PartdandeofProblem9.5
% Problem Set 9, Problem 1: Round Rotor Machine
xd = 2; % synchronous reactance
Pb = 1e9; % rated power
Vb = 24000; % rated terminal voltage, line-line, RMS
Ifnl = 1000; % field current for unity internal voltage
Ib = Pb/(sqrt(3)*Vb); % base current
pf = .8; % can reach this power factor at rated VA
% Part a:
psi = acos(pf); % this is the rated power factor angle
eaf = sqrt(1+xd^2+2*xd*sin(psi)); % law of cosines calculation
vr = xd*sin(psi);
vi = xd*cos(psi);
fprintf(eaf = %g vr = %g vi = %g\n, eaf, vr, vi)
eang = (180/pi)*atan(vi/(1+vr));
vang = (180/pi)*atan(vi/vr);
fprintf(angles: eaf = %g vx = %g\n, eang, vang)
4
1000 MVA
Field Current Limit
Armature Current Limit
Stability Limit
Figure5: CapabilityCurveforProblem9.5
% part b:
eafmin = pf*xd; % this will be the stability limit
Eaf = eafmin:.01:eaf; % to generate a plot
Delta = asin(xd*pf./ Eaf);
Deltad = (180/pi) .* Delta;
figure(1)
plot(Eaf, Deltad)
title(Problem 9.5, part b)
ylabel(Degrees)
xlabel(Per-Unit E_af)
grid on
% part c:
ppu = Eaf .* sin(Delta) ./ xd;
qpu = Eaf .* cos(Delta) ./ xd -1/xd;
I_a = Ib .* sqrt(ppu .^2 + qpu .^2);
figure(2)
plot(Eaf, I_a)
5
title(Problem 9.5, part b)
ylabel(|I_a|)
xlabel(Per-Unit E_{af})
grid on
% pard d and e: generate vee curve
% first, get zero power curve: need only corners
iamax = (eaf-1)/xd;
iamin = 1/xd;
I_az = Ib .* [iamin 0 iamax];
E_az = Ifnl .* [0 1 eaf];
figure(3)
clf
hold on
plot(E_az, I_az)
P_pu = [.2 .4 .6 .8];
for k = 1:length(P_pu)
p = P_pu(k);
deltm = asin(xd*p/eaf); % delta for the top of each curve
Delt = deltm:.01:pi/2; % from there to stability limit
E = p*xd ./sin(Delt); % required per-unit excitation
ppu = E .* sin(Delt) ./ xd;
qpu = E .* cos(Delt) ./ xd - 1/xd;
I_a = Ib .* sqrt(ppu .^2 + qpu .^2);
I_f = Ifnl .* E;
plot(I_f, I_a)
end
% now, just to be fancy, we generate the limits
ylim = Ib .* [1 1 0];
xlim = Ifnl*eaf .* [0 1 1];
plot(xlim, ylim, --)
hold off
grid on
title(Problem 9.5, part e)
ylabel(Armature Current Magnitude, A RMS)
xlabel(Field Current, A)
Problem 2: Chapter 9, Problem 6 Revisited Here wend the capability curve for this salient
polemachine. Aswiththeroundrotormachine,wehavethreebasiclimitstocapability:
6
Limit 3
p
q
Point B
Point A
Limit 1
Limit 2
1. Field Current Limit (heating)
2. Armature Current Limit (heating)
3. UnderexcitedStabilityLimit
TheseareshowninFigure. Thethreelimitsarecalledoutwiththeirnumbersasabove.
There are two interesting points thatare the places where two limts meet: Point A is the point
thateldandarmaturecurrentlimitshaveincommon. Itisalsothepointthatestablishes
machinerating. PointBiswherearmaturecurrentlimitisalsoatthestabilitylimit. This
problem,too,iscarriedoutusingthescriptappended.
Tostart,usethevectordiagramforsalientpolemotoroperationshowninFigure6,whichis
alsoFigure9.10ofthetext.
Point A
Assumingthis isoperatingatratedarmaturecurrent (Per-uniti= 1),ratedinternalvoltage
andtorqueangleare:
= cos
1
PF
= icos +jisin
e
1
= vjx
q
i
=

e
1
7
i



1
V
I
j X I
d d
j X I
q q
j X I
q
q axis
d axis
I
d
I
q
af
E
E
Figure6: VectorDiagremforSalientPoleMachine
i
d
= isin ()
e
af
= |e
1
+(x
d
x
q
) i
d
Actually,whatiscalculatedisactuallythenegativeofi
d
, butthat is justa little bitmore
convenient.
Point B
Thisistheintersectionofthestabilitylimit,where
x
d
cos 2
e
af
=v 1
x
q
cos
andthearmaturecurrentlimit,wherep+jq=1;
Asitturnsout,thiscanbefoundbysearchingovertheangle and,calculatinge
af
foreach
valueof,ndingwherearmaturecurrentisrated.
Thesetwopointsgiveustheappropriatelimits. Thenforsegment1,wesetupavectorof
closelyspacedpointsinanglefrom
A
< <0. (Notethat
A
isnegative.) Internalvoltage
isconstant,andrealandreactivepowerarecalculated.
Forsegment2,between
B
< <
A
,itisstraightforwardtosetupavectorofcloselyspaced
pointsinpowerfactorangle,andp= cos and q= sin .
Thelastsegmentisactuallymostinteresting.Asitturnsout,theangledoesgobacktozero
and internal eld voltage crosses zero and becomes negative. The stability limit is as we
8
calculateditabove,andthatcanbecalculatedforeachofacloselyspacedvectorofpoints:

B
< <0. Realandreactivepowerarecalculatedateachpoint.
Togeneratetheplot,thethreesegmentsareplottedseparately,andtheresultisshownin
Figure 7.
1
0.5
0
0.5
0.2 0 0.2 0.4 0.6 0.8 1 1.2 1.4
Field Limit
Armature Limit
Stability Limit
Figure 7: CapabilityCurve for Salient Pole Machine (per-unit)
Script for Problem 2
% Capability Curve for a Salient Generator
global xd xq v eq d
%per-unit
xd = 1.5;
xq = 1.0;
v=1;
Pb = 1e6;
Vb = 4200/sqrt(3);
Ib = Pb/(3*Vb);
Ifnl = 100;
pf0 = 1.0;
% step 1: find rated eaf and angle at rating point (Point A)
psi0 = acos(pf0);
Im = 1;
I = Im * (cos(psi0) + j*sin(psi0));
e1 = v - j*xq*I;
9
delt0 = angle(e1);
I_d = Im * sin(psi0-delt0); % actually -I_d
eaf0 = abs(e1) + I_d * (xd - xq);
fprintf(Point A: eaf = %g delta = %g\n, eaf0, delt0)
% step 2 find point B Intersection of rated current and stability
delt1 = fzero(ff, [-pi/2 0]);
eaf1 = v*(1-xd/xq)*cos(2*delt1)/cos(delt1);
fprintf(Point B: eaf = %g delta = %g\n, eaf1, delt1)
p1 = -.5*v^2*(1/xq-1/xd)*sin(2*delt1)-v*eaf1*sin(delt1)/xd;
q1 = .5*v^2*(1/xq+1/xd) - .5*v^2*(1/xq-1/xd)*cos(2*delt1) - v*eaf1*cos(delt1)/xd;
psi1 = atan(q1/p1);
fprintf(Point B: P = %g Q = %g psi = %g\n, p1, q1, psi1)
% now run it out
% region 1 is constant eaf: field limit
Delt1 = delt0:.01:0;
eaf = eaf0;
P1 = -.5*v^2*(1/xq-1/xd) .* sin(2 .* Delt1)...
-(v*eaf/xd) .* sin(Delt1);
Q1 = .5*v^2*(1/xq+1/xd) -.5*v^2*(1/xq-1/xd) .* cos(2 .* Delt1) ...
-(v*eaf/xd) .* cos(Delt1);
% region 2 is constant armature current magnitude
psi2 = -psi0:.01:psi1;
P2 = cos(psi2);
Q2 = sqrt(1-P2 .^2);
% region 3 is the stability limit
Delt3 = delt1:.01:0;
eaf3 = v*(1-xd/xq)*cos(2 .* Delt3) ./ cos(Delt3);
P3 = -.5*v^2*(1/xq-1/xd) .* sin(2 .* Delt3)...
-(v .*eaf3 ./xd) .* sin(Delt3);
Q3 = .5*v^2*(1/xq+1/xd) -.5*v^2*(1/xq-1/xd) .* cos(2 .* Delt3) ...
-(v .*eaf3 ./xd) .* cos(Delt3);
10
---------------
L1x = [0 0];
L1y = [-.5 1];
L2x = [0 1.2];
L2y = [0 0];
figure(1)
plot(P1, Q1, P2, Q2, P3, Q3, L1x, L1y, L2x, L2y)
axis square
axis equal
grid on
legend(Field Limit, Armature Limit, Stability Limit)
function deltb = ff(delt)
global xd xq v eq d
eaf = v*(1-xd/xq)*cos(2*delt)/cos(delt);
p = -.5*v^2*(1/xq-1/xd)*sin(2*delt)-v*eaf*sin(delt)/xd;
q = .5*v^2*(1/xq+1/xd) - .5*v^2*(1/xq-1/xd)*cos(2*delt) - v*eaf*cos(delt)/xd;
deltb = 1-abs(p+j*q);
Problem 3 With48slotsandafourpolemachine,thenumberofslotsperpoleperphaseis:
48
m= = 4
6 2
360
Andtheelectrical anglebetweenslotsis = 2
48
= 15

. Thebreadthfactoris:
sin
m
sin30

k
b
=
2
= .957
msin

4 sin 7.5

2
Forthefullpitchwinding,ofcourse,k
p
=1. Shortpitchingthewindingreducesthepitch
anglebythenumberofslotstimes,sothat,foroneslotshorttheangleis= 165

and for
twoshortitis= 150

. So:
full k
w
= .957 1.0 .957
1short k
w
= .957 .991 .948
2short k
w
= .957 .966 .924
In this context we refer to the one slot short pitchedwindingas11/12andthe two slot short
pitchas5/6.
11
MIT OpenCourseWare
http://ocw.mit.edu
6.061 / 6.690 Introduction to Electric Power Systems
Spring 2011
For information about citing these materials or our Terms of Use, visit: http://ocw.mit.edu/terms.