SYNCHRONOUS MACHINES




Copyright © P. Kundur
This material should not be used without the author's consent
                                                                1539pk
Synchronous Machines

                      Outline

1.   Physical Description
2.   Mathematical Model
3.   Park's "dqo" transportation
4.   Steady-state Analysis
        phasor representation in d-q coordinates
        link with network equations
1.   Definition of "rotor angle"
2.   Representation of Synchronous Machines in
     Stability Studies
          neglect of stator transients
          magnetic saturation
1.   Simplified Models
2.   Synchronous Machine Parameters
3.   Reactive Capability Limits




                            SM - 2
                                                    1539pk
Physical Description of a
            Synchronous Machine

   Consists of two sets of windings:
       3 phase armature winding on the stator
        distributed with centres 120° apart in space
       field winding on the rotor supplied by DC
   Two basic rotor structures used:
       salient or projecting pole structure for hydraulic
        units (low speed)
       round rotor structure for thermal units (high
        speed)
   Salient poles have concentrated field windings;
    usually also carry damper windings on the pole
    face.
    Round rotors have solid steel rotors with
    distributed windings
   Nearly sinusoidal space distribution of flux wave
    shape obtained by:
       distributing stator windings and field windings in
        many slots (round rotor);
       shaping pole faces (salient pole)



                           SM - 3
                                                        1539pk
Rotors of Steam Turbine Generators


   Traditionally, North American manufacturers normally
    did not provide special “damper windings”
       solid steel rotors offer paths for eddy currents,
        which have effects equivalent to that of amortisseur
        currents
   European manufacturers tended to provide for
    additional damping effects and negative-sequence
    current capability
       wedges in the slots of field windings
        interconnected to form a damper case, or
       separate copper rods provided underneath the
        wedges




          Figure 3.3: Solid round rotor construction

                             SM - 4
                                                       1539pk
Rotors of Hydraulic Units


   Normally have damper windings or amortisseurs
       non-magnetic material (usually copper) rods
        embedded in pole face
       connected to end rings to form short-circuited
        windings
   Damper windings may be either continuous or non-
    continuous
   Space harmonics of the armature mmf contribute to
    surface eddy current
       therefore, pole faces are usually laminated




         Figure 3.2: Salient pole rotor construction


                             SM - 5
                                                         1539pk
Balanced Steady State Operation


   Net mmf wave due to the three phase stator
    windings:
       travels at synchronous speed
       appears stationary with respect to the rotor; and
       has a sinusoidal space distribution

   mmf wave due to one phase:




         Figure 3.7: Spatial mmf wave of phase a



                           SM - 6
                                                      1539pk
Balanced Steady State Operation


   The mmf wave due to the three phases are:



    MMFa = Kia cos γ                  ia = Im cos( ωs t )

                       2π                              2π 
    MMFb = Kib cos γ −              ib = Im cos ωs t −    
                        3                               3 

                       2π                              2π 
    MMFc = Kic cos γ +              ia = lm cos ωs t +    
                       3                               3 




           MMFtotal = MMFa + MMFb + MMFc

                           3
                       =     KIm cos( γ − ωs t )
                           2




                                 SM - 7
                                                                 1539pk
Balanced Steady State Operation


   Magnitude of stator mmf wave and its relative
    angular position with respect to rotor mmf wave
    depend on machine output
       for generator action, rotor field leads stator field
        due to forward torque of prime mover;
       for motor action rotor field lags stator field due
        to retarding torque of shaft load




      Figure 3.8: Stator and rotor mmf wave shapes




                            SM - 8
                                                         1539pk
Transient Operation


   Stator and rotor fields may:
       vary in magnitude with respect to time
       have different speed
   Currents flow not only in the field and stator
    windings, but also in:
       damper windings (if present); and
       solid rotor surface and slot walls of round rotor
        machines




         Figure 3.4: Current paths in a round rotor



                            SM - 9
                                                       1539pk
Direct and Quadrature Axes


    The rotor has two axes of symmetry
    For the purpose of describing synchronous
     machine characteristics, two axes are defined:
        the direct (d) axis, centered magnetically in the
         centre of the north pole
        The quadrature (q) axis, 90 electrical degrees
         ahead of the d-axis




    Figure 3.1: Schematic diagram of a 3-phase synchronous
                            machine

                            SM - 10
                                                          1539pk
Mathematical Descriptions of a
           Synchronous Machine

   For purposes of analysis, the induced currents in
    the solid rotor and/or damper windings may be
    assumed to flow in two sets of closed circuits
       one set whose flux is in line with the d-axis; and
       the other set whose flux is along the q-axis
   The following figure shows the circuits involved




           Figure 3.9: Stator and rotor circuits



                            SM - 11
                                                       1539pk
Review of Magnetic Circuit Equations
          (Single Excited Circuit)

   Consider the elementary circuit of Figure 3.10
                               dΨ
                        ei =
                               dt
                               dΨ
                        e1 =      + ri
                               dt
                        Ψ = Li

   The inductance, by definition, is equal to flux linkage
    per unit current
                               φ
                       L=N       = N2P
                               i
    where
       P = permeance of magnetic path
        = flux = (mmf) P = NiP




        Figure 3.10: Single excited magnetic circuit
                               SM - 12
                                                       1539pk
Review of Magnetic Circuit Equations
              (Coupled Circuits)

   Consider the circuit shown in Figure 3.11
                            dΨ1
                     e1 =       + r1i1
                             dt
                            dΨ2
                     e2 =       + r2i2
                             dt
                     Ψ1 = L11i1 + L21i2

                     Ψ2 = L21i1 + L22i2

    with L11 = self inductance of winding 1
         L22 = self inductance of winding 2
         L21 = mutual inductance between winding 1 and 2




           Figure 3.11: Magnetically coupled circuit

                                SM - 13
                                                       1539pk
Basic Equations of a Synchronous Machine

   The equations are complicated by the fact that the
    inductances are functions of rotor position and
    hence vary with time
   The self and mutual inductances of stator circuits
    vary with rotor position since the permeance to flux
    paths vary
                Iaa = L al + Igaa

                    = L aa 0 + L aa 2 cos 2θ

                                                      2π 
                Iab = Iba = −L ab 0 + L ab 2 cos 2θ −    
                                                       3 
                                                  π
                    = −L ab 0 − L ab 2 cos 2θ + 
                                                 3

   The mutual inductances between stator and rotor
    circuits vary due to relative motion between the
    windings



                  Iafd = L afd cos θ

                  Iakd = L akd cos θ

                                      π
                  Iakq = L akq cos θ +  = −L akq sin θ
                                      2

                                    SM - 14
                                                              1539pk
Basic Equations of a Synchronous Machine

    Dynamics of a synchronous machine is given by the
     equations of the coupled stator and rotor circuits
    Stator voltage and flux linkage equations for phase a
     (similar equations apply to phase b and phase c)
                      dΨa
               ea =       − R aia = pΨa − R aia
                       dt
               Ψa = −laaia −labib − lacic + lafdifd + lakdikd + lakqikq

    Rotor circuit voltage and flux linkage equations
                                    e fd = pΨfd + R fdifd

                                      0 = pΨkd + R kdikd

                                      0 = pΨkq + R kqikq
       ψ fd = L ffdifd + L fkdikd
                                          2π               2π  
            − L afd ia cos θ + ib cos θ −     + ic cos θ +    
                                          3                 3 
       ψ kd = L fkdifd + Lkkdikd
                                          2π               2π  
            − L afd ia cos θ + ib cos θ −     + ic cos θ +    
                                          3                 3 
       ψ kq = Lkkdikq
                                          2π               2π  
            + L akq ia sin θ + ib sin θ −     + ic sin θ +    
                                          3                 3 

                                            SM - 15
                                                                          1539pk
The dqo Transformation

   The dqo transformation, also called Park's
    transformation, transforms stator phase quantities from
    the stationary abc reference frame to the dqo reference
    frame which rotates with the rotor
                                    2π           2π  
                  cos θ     cos θ −      cos θ +    
                                     3            3 
         id                                                 ia 
         i  = 2                   2π           2π     ib 
           3
           q      − sin θ − sin θ − 3  − sin θ + 3  
                                                          i 
         i0     1                   1               1      c
                                                         
                  2
                                      2               2 

    The above transformation also applies to stator flux
    linkages and voltages

   With the stator quantities expressed in the dqo reference
    frame
       all inductances are independent of rotor position
        (except for the effects of magnetic saturation)
       under balanced steady state operation, the stator
        quantities appear as dc quantities
       during electromechanical transient conditions,
        stator quantities vary slowly with frequencies in
        the range of 1.0 to 3.0 Hz
    The above simplify computation and analysis of results.

                                   SM - 16
                                                                      1539pk
Physical Interpretation of dqo
               Transformation


   The dqo transformation may be viewed as a means
    of referring the stator quantities to the rotor side
   In effect, the stator circuits are represented by two
    fictitious armature windings which rotate at the
    same speed as the rotor; such that:
       the axis of one winding coincides with the d-axis
        and that of the other winding with the q-axis
       The currents id and iq flowing in these circuits
        result in the same mmf's on the d- and q-axis as
        do the actual phase currents
   The mmf due to id and iq are stationary with respect
    to the rotor, and hence:
       act on paths of constant permeance, resulting in
        constant self inductances (Ld, Lq) of stator
        windings
       maintain fixed orientation with rotor circuits,
        resulting in constant mutual inductances




                            SM - 17
                                                          1539pk
Per Unit Representation


   The per unit system is chosen so as to further
    simplify the model
   The stator base quantities are chosen equal to the
    rated values
   The rotor base quantities are chosen so that:
       the mutual inductances between different
        circuits are reciprocal (e.g. Lafd = Lfda)

       the mutual inductances between the rotor and
        stator circuits in each axis are equal (e.g., Lafd =
        Lakd)
          The P.U. system is referred to as the "Lad
          base reciprocal P.U. system"

   One of the advantages of having a P.U. system with
    reciprocal mutual inductances is that it allows the
    use of equivalent circuits to represent the
    synchronous machine characteristics




                            SM - 18
                                                           1539pk
P.U. Machine Equations in
             dqo reference frame


   The equations are written with the following
    assumptions and notations:
       t is time in radians
       p = d/dt
       positive direction of stator current is out of the
        machine
       each axis has 2 rotor circuits
   Stator voltage equations
              e d = pψ d − ψ qωr − R aid

              e q = pψ q + ψ dωr − R aiq

              e 0 = pψ 0 − R a i 0

   Rotor voltage equations
                e fd = pψ fd + R fdifd

                0 = pψ 1d + R 1di1d

                0 = pψ 1q + R 1qi1q

                0 = pψ 2q + R 2qi2q


                                SM - 19
                                                         1539pk
P.U. Machine Equations in dqo Reference
             Frame (cont'd)

   Stator flux linkage equations
        ψ d = −( Lad + Ll ) id + Lad ifd + Lad i1d
        ψ q = − ( Laq + Ll ) iq + Laq i1q + Laq i 2 q

        ψ 0 = −L0 i0

   Rotor flux linkage equations

            ψ fd = L ffdifd + L f 1di1d − L adid

            ψ 1d = L f 1difd + L11di1d − L adid

            ψ 1q = L11qi1q + L aqi2q − L aqiq

            ψ 1q = L aqi1q +L 22qL 2 q − L aqiq


   Air-gap torque

                 T e = ψ diq − ψ qid




                                 SM - 20
                                                        1539pk
Steady State Analysis Phasor
                 Representation

For balanced, steady state operation, the stator voltages may
be written as:
              e a = Em cos( ωt + α )

              eb = Em cos( ωt − 2π 3 + α )

              e c = Em cos( ωt + 2π 3 + α )
with
       ω = angular velocity = 2πf
       α = phase angle of ea at t=0


Applying the d,q transformation,

               e d = Em cos( ωt + α − θ )

               e q = Em sin( ωt + α − θ )

At synchronous speed, the angle θ is given by θ = ωt + θ0
with θ = value of θ at t = 0


Substituting for θ in the expressions for ed and eq,
                   e d = Em cos( α − θ 0 )

                   e q = Em sin( α − θ 0 )

                                  SM - 21
                                                        1539pk
Steady State Analysis Phasor
              Representation (cont'd)

   The components ed and eq are not a function of t because
    rotor speed ω is the same as the angular frequency ω
    of the stator voltage. Therefore, ed and eq are constant
    under steady state.

    In p.u. peak value Em is equal to the RMS value of terminal
    voltage Et. Hence,
                       e d = E t cos( α −θ0 )

                       e q = E t sin( α −θ0 )

   The above quantities can be represented as phasors with
    d-axis as real axis and q-axis as imaginary axis




    Denoting δi, as the angle by which q-axis leads E
                         e d = E t sin δi

                         e q = E t cos δi

                                 SM - 22
                                                            1539pk
Steady State Analysis Phasor
           Representation (cont'd)

   The phasor terminal voltage is given by

               ~               in the d-q coordinates
               E t = ed + je q
                                 in the R-I coordinates
                  = E R + jE l
   This provides the link between d,q components in a
    reference frame rotating with the rotor and R, I
    components associated with the a.c. circuit theory

   Under balanced, steady state conditions, the d,q,o
    transformation is equivalent to
       the use of phasors for analyzing alternating
        quantities, varying sinusoidally with respect to
        time

   The same transformation with θ = ωt applies to both
       in the case of machines, ω = rotor speed
       in the case of a.c. circuits, ω = angular frequency




                            SM - 23
                                                          1539pk
Internal Rotor Angle

   Under steady state
                 e d = −ωψq − idR a

                   = ωL qiq − idR a = X qiq −idR a
    Similarly
                 e q = ωψd − iqR a

                   = − X did + X adifd − iqR a
   Under no load, id=iq=0. Therefore,

                       ψ q = −L qiq = 0

                       ψ d = L adifd

                        ed = 0

                        e q = L adifd

        ~
    and E t = e d + jeq = jL adifd


   Under no load, Et has only the q-axis component
    and δi=0. As the machine is loaded, δi increases.

    Therefore, δi is referred to as the load angle or
    internal rotor angle.
   It is the angle by which q-axis leads the phasor Et
                                 SM - 24
                                                        1539pk
Electrical Transient Performance


   To understand the nature of electrical transients, let
    us first consider the RL circuit shown in Figure 3.24
    with e = Emsin (ωt+α). If switch "S" is closed at t=0,
    the current is given by
                                      di
                             e=L         + iR
                                      dt
    solving
                        ( L) t       Em
                                        sin( ωt + α − φ)
                        −R
               i = Ke            +
                                     Z
   The first term is the dc component. The presence of
    the dc component ensures that the current does not
    change instantaneously. The dc component decays
    to zero with a time constant of L/R




                     Figure 3.24: RL Circuit

                                     SM - 25
                                                           1539pk
Short Circuit Currents of a Synchronous
                    Machine

    If a bolted three-phase fault is suddenly applied to
     a synchronous machine, the three phase currents
     are shown in Figure 3.25.




         Figure 3.25: Three-phase short-circuit currents

                              SM - 26
                                                            1539pk
Short Circuit Currents of a Synchronous
                 Machine (cont'd)

    In general, fault current has two distinct
     components:
       a) a fundamental frequency component which
          decays initially very rapidly (a few cycles) and
          then relatively slowly (several seconds) to a
          steady state value
       b) a dc component which decays exponentially in
          several cycles
    This is similar to the short circuit current in the case
     of the simple RL circuit. However, the amplitude of
     the ac component is not constant
        internal voltage, which is a function of rotor flux
         linkages, is not constant
        the initial rapid decay is due to the decay of flux
         linking the subtransient circuits (high resistance)
        the slowly decaying part of the ac component is
         due to the transient circuit (low resistance)
    The dc components have different magnitudes in
     the three phases




                            SM - 27
                                                             1539pk
Elimination of dc Component by
        Neglecting Stator Transients

   For many classes of problems, considerable
    computational simplicity results if the effects of ac
    and dc components are treated separately
   Consider the stator voltage equations
                e d = pψ d − ω ψ q − idR a
                e q = pψ q + ω ψ d − iqR a

    transformer voltage terms: pψd, pψq

    speed voltage terms: ω ψ q , ω ψ d


   The transformer voltage terms represent stator
    transients:
       stator flux linkages (ψd, ψq) cannot change
        instantaneously
       result in dc offset in stator phasor current
   If only fundamental frequency stator currents are of
    interest, stator transients (pψd, pψq) may be
    neglected.



                             SM - 28
                                                       1539pk
Short Circuit Currents with Stator
             Transients Neglected

   The resulting stator phase currents following a
    disturbance has the wave shape shown in Figure
    3.27
   The short circuit has only the ac component whose
    amplitude decays
   Regions of subtransient, transient and steady state
    periods can be readily identified from the wave shape
    of phase current




     Figure 3.27: Fundamental frequency component of short
                     circuit armature current


                             SM - 29
                                                             1539pk
Synchronous Machine Representation in
           System Stability Studies

    Stator Transients (pψd, pψq) are usually neglected

        accounts for only fundamental frequency
         components of stator quantities

        dc offset either neglected or treated separately

        allows the use of steady-state relationships for
         representing the transmission network



    Another simplifying assumption normally made is
     setting ω = 1 in the stator voltage equations

        counter balances the effect of neglecting stator
         transients so far as the low-frequency rotor
         oscillations are concerned

        with this assumption, in per unit air-gap power
         is equal to air-gap torque



       (See section 5.1 of book for details)


                           SM - 30
                                                          1539pk
Equation of Motion (Swing Equation)


   The combined inertia of the generator and prime-
    mover is accelerated by the accelerating torque:

                     dωm
                 J       = Ta = Tm − Te
                      dt

    where

    Tm =    mechanical torque in N-M

    Te =    electromagnetic torque in N-m

    J   =   combined moment of inertia of generator
            and turbine, kg•m2

    m =    angular velocity of the rotor in mech. rad/s

    t   =   time in seconds




                             SM - 31
                                                       1539pk
Equation of Motion (cont'd)


   The above equation can be normalized in terms of
    per unit inertia constant H
                           1 Jω2 m
                        H=      0
                           2 VA base
    where
        0m = rated angular velocity of the rotor in
       mechanical radians per second
   Equation of motion in per unit form is
                          d ωr
                     2H        = Tm − Te
                           dt
    where
        ωm
ωr =              = per unit rotor angular velocity
        ω0m
       Tmω0m
Tm =
       VA base    = per unit mechanical torque

       Te ω0m
Te =
       VAbase     = per unit electromechanical torque

   Often inertia constant M = 2H used

                                 SM - 32
                                                        1539pk
Magnetic Saturation


   Basic equations of synchronous machines
    developed so far ignored effects of saturation
       analysis simple and manageable
       rigorous treat a futile exercise
   Practical approach must be based on semi-
    heuristic reasoning and judiciously chosen
    approximations
       consideration to simplicity, data availability,
        and accuracy of results
   Magnetic circuit data essential to treatment of
    saturation given by the open-circuit characteristic
    (OCC)




                           SM - 33
                                                          1539pk
Assumptions Normally Made in the
       Representation of Saturation

   Leakage inductances are independent of saturation
   Saturation under loaded conditions is the same as
    under no-load conditions
   Leakage fluxes do not contribute to iron saturation
       degree of saturation determined by the air-gap
        flux
   For salient pole machines, there is no saturation in
    the q-axis
       flux is largely in air
   For round rotor machines, q-axis saturation
    assumed to be given by OCC
       reluctance of magnetic path assumed
        homogeneous around rotor periphery




                            SM - 34
                                                      1539pk
   The effects of saturation is represented as

                       L ad = K sdL adu            (3.182)

                       L aq = K sqL aqu            (3.183)


    Ladu and Laqu are unsaturated values. The saturation
    factors Ksd and Ksq identify the degrees of saturation.
   As illustrated in Figure 3.29, the d-axis saturation is
    given by The OCC.
   Referring to Figure 3.29,


                        ΨI = Ψat 0 − Ψat           (3.186)

                                   Ψat             (3.187)
                        K sd =
                                 Ψat + ΨI
   For the nonlinear segment of OCC,     can be
                                      ΨI
    expressed by a suitable mathematical function:


                  ΨI = A sat eBsat ( Ψat − ΨTI )     (3.189)




                                 SM - 35
                                                               1539pk
Open-Circuit Characteristic (OCC)


     Under no load rated speed conditions
                 id = iq = Ψq = e d = 0

                   E t = e q = Ψd = L adifd
     Hence, OCC relating to terminal voltage and field
      current gives saturation characteristic of the d-axis




    Figure 3.29: Open-circuit characteristic showing effects of
                           saturation


                                 SM - 36
                                                                  1539pk
   For salient pole machines, since q-axis flux is
    largely in air, Laq does not vary significantly with
    saturation
       Ksq=1 for all loading conditions
   For round rotor machines, there is saturation in
    both axes
       q-axis saturation characteristic not usually
        available
       the general industry practice is to assume
        Ksq = Ksd
   For a more accurate representation, it may be
    desirable to better account for q-axis saturation of
    round rotor machines
       q-axis saturates appreciably more than the d-
        axis, due to the presence of rotor teeth in the
        magnetic path
   Figure 3.32 shows the errors introduced by
    assuming q-axis saturation to be same as that of
    d-axis, based on actual measurements on a 500
    MW unit at Lambton GS in Ontario
       Figure shows differences between measured
        and computed values of rotor angle and field
        current
       the error in rotor angle is as high as 10%, being
        higher in the underexcited region
       the error in the field current is as high as 4%,
        being greater in the overexcited region
                             SM - 37
                                                            1539pk
    The q-axis saturation characteristic is not readily
     available
        It can, however, be fairly easily determined from
         steady-state measurements of field current and
         rotor angle at different values of terminal voltage,
         active and reactive power output
        Such measurements also provide d-axis
         saturation characteristics under load
        Figure 3.33 shows the d- and q-axis saturation
         characteristics derived from steady-state
         measurements on the 500 MW Lambton unit




     Figure 3.33: Lambton saturation curves derived from
    steady-state field current and rotor angle measurements


                               SM - 38
                                                              1539pk
Example 3.3
   Considers the 555 MVA unit at Lambton GS and
    examines
       the effect of representing q-axis saturation
        characteristic distinct from that of d-axis
       the effect of reactive power output on rotor angle


   Table E3.1 shows results with q-axis saturation assumed
    same as d-axis saturation
                            Table E3.1
       Pt        Qt      Ea (pu)             Ksd   δi (deg)   ifd (pu)
       0         0        1.0            0.889        0       0.678
      0.4       0.2      1.033           0.868      25.3      1.016
      0.9      0.436     1.076           0.835      39.1      1.565
      0.9        0       1.012           0.882      54.6      1.206
      0.9       -0.2     0.982           0.899      64.6      1.089

   Table E3.2 shows results with distinct d- and q-axis
    saturation representation
                            Table E3.2
       Pt        Qt        Ksq               Ksd   δi (deg)   ifd (pu)
       0         0       0.667           0.889        0       0.678
      0.4       0.2      0.648           0.868      21.0      1.013
      0.9      0.436     0.623           0.835      34.6      1.559
      0.9        0       0.660           0.882      47.5      1.194
      0.9       -0.2     0.676           0.899      55.9      1.074

                                   SM - 39
                                                                         1539pk
Simplified Models for Synchronous
                Machines

   Neglect of Amortisseurs
       first order of simplification
       data often not readily available



   Classical Model (transient performance)
       constant field flux linkage
       neglect transient saliency (x'd = x'q)
                                       Et
                            x′d
                E´




   Steady-state Model
       constant field current
       neglect saliency (xd = xq = xs)

                                  Et
                       xs               Eq = Xadifd
           Eq




                            SM - 40
                                                      1539pk
Reactive Capability Limits of Synchronous
               Machines

   In voltage stability and long-term stability studies, it
    is important to consider the reactive capability
    limits of synchronous machines
   Synchronous generators are rated in terms of
    maximum MVA output at a specified voltage and
    power factor which can be carried continuously
    without overheating
   The active power output is limited by the prime
    mover capability
   The continuous reactive power output capability is
    limited by three considerations
       armature current limit

       field current limit

       end region heating limit




                              SM - 41
                                                         1539pk
Armature Current Limit


   Armature current results in power loss, and the
    resulting heat imposes a limit on the output

    The per unit complex output power is
                     ~ *
        S = P + jQ = E t ~t = E t It ( cos φ + j sin φ )
                         I

    where Φis the power factor angle

   In a P-Q plane the armature current limit, as shown
    in Fig. 5.12, appears as a circle with centre at the
    origin and radius equal to the MVA rating




          Fig 5.12: Armature current heating limit


                                 SM - 42
                                                           1539pk
Field Current Limit


   Because of the heating resulting from RfdI2fd power loss,
    the field current imposes the second limit
   The phasor diagram relating Et, It and Eq (with Ra
    neglected) is shown in Fig. 5.13

    Equating the components along and perpendicular to
    the phasor Et
                   X adifd sin δ i = X slt cos φ

                   X adifd cos δ i = E t + X slt sin φ

    Therefore
                                  X ad
                 P = E tlt cos φ =     E tifd sin δi
                                   Xs
                                  X                  E2
                 Q = E tlt sin φ = ad E tifd cos δi − t
                                   Xs                Xs

   The relationship between P and Q for a given field
    current is a circle centered at on the Q-axis and with
    as the radius. The effect of the maximum field current
    on the capability of the machine is shown in Fig. 5.14
   In any balanced design, the thermal limits for the field
    and armature intersect at a point (A) which represents
    the machine name-plate MVA and power factor rating



                                SM - 43
                                                          1539pk
Field Current Limit




Fig. 5.13: Steady state phasor diagram




Fig. 5.14: Field current heating limit
                 SM - 44
                                         1539pk
End Region Heating Limit


   The localized heating in the end region of the armature
    affects the capability of the machine in the underexcited
    condition
   The end-turn leakage flux, as shown in Fig. 5.15, enters
    and leaves in a direction perpendicular (axial) to the
    stator lamination. This causes eddy currents in the
    laminations resulting in localized heating in the end
    region
   The high field currents corresponding to the overexcited
     condition keep the retaining ring saturated, so that end
    leakage flux is small. However, in the underexcited
    region the field current is low and the retaining ring is
    not saturated; this permits an increase in armature end
    leakage flux
   Also, in the underexcited condition, the flux produced
    by the armature current adds to the flux produced by the
    field current. Therefore, the end-turn flux enhances the
    axial flux in the end region and the resulting heating
    effect may severely limit the generator output,
    particularly in the case of a round rotor machine
   Fig. 5.16 shows the locus of end region heating limit on
    a P-Q plane



                            SM - 45
                                                         1539pk
End Region Heating Limit




Fig. 5.15: Sectional view end region of a generator




         Fig. 5.16: End region heating limit


                        SM - 46
                                                      1539pk
Reactive Capability Limit of a 400 MVA
Hydrogen Cooled Steam Turbine Generator

    Fig. 5.18 shows the reactive capability curves of a 400
     MVA hydrogen cooled steam turbine driven generator
     at rated armature voltage
         the effectiveness of cooling and hence the
          allowable machine loading depends on hydrogen
          pressure
         for each pressure, the segment AB represents the
          field heating limit, the segment BC armature
          heating limit, and the segment CD the end region
          heating limit




    Fig. 5.18: Reactive capability curves of a hydrogen cooled
                    generator at rated voltage
                               SM - 47
                                                             1539pk
Effect of Changes in Terminal Voltage Et




 Fig. 5.17: Effect of reducing the armature voltage on the
                 generator capability curve



                           SM - 48
                                                             1539pk

02 1 synchronous-machines

  • 1.
    SYNCHRONOUS MACHINES Copyright ©P. Kundur This material should not be used without the author's consent 1539pk
  • 2.
    Synchronous Machines Outline 1. Physical Description 2. Mathematical Model 3. Park's "dqo" transportation 4. Steady-state Analysis  phasor representation in d-q coordinates  link with network equations 1. Definition of "rotor angle" 2. Representation of Synchronous Machines in Stability Studies  neglect of stator transients  magnetic saturation 1. Simplified Models 2. Synchronous Machine Parameters 3. Reactive Capability Limits SM - 2 1539pk
  • 3.
    Physical Description ofa Synchronous Machine  Consists of two sets of windings:  3 phase armature winding on the stator distributed with centres 120° apart in space  field winding on the rotor supplied by DC  Two basic rotor structures used:  salient or projecting pole structure for hydraulic units (low speed)  round rotor structure for thermal units (high speed)  Salient poles have concentrated field windings; usually also carry damper windings on the pole face. Round rotors have solid steel rotors with distributed windings  Nearly sinusoidal space distribution of flux wave shape obtained by:  distributing stator windings and field windings in many slots (round rotor);  shaping pole faces (salient pole) SM - 3 1539pk
  • 4.
    Rotors of SteamTurbine Generators  Traditionally, North American manufacturers normally did not provide special “damper windings”  solid steel rotors offer paths for eddy currents, which have effects equivalent to that of amortisseur currents  European manufacturers tended to provide for additional damping effects and negative-sequence current capability  wedges in the slots of field windings interconnected to form a damper case, or  separate copper rods provided underneath the wedges Figure 3.3: Solid round rotor construction SM - 4 1539pk
  • 5.
    Rotors of HydraulicUnits  Normally have damper windings or amortisseurs  non-magnetic material (usually copper) rods embedded in pole face  connected to end rings to form short-circuited windings  Damper windings may be either continuous or non- continuous  Space harmonics of the armature mmf contribute to surface eddy current  therefore, pole faces are usually laminated Figure 3.2: Salient pole rotor construction SM - 5 1539pk
  • 6.
    Balanced Steady StateOperation  Net mmf wave due to the three phase stator windings:  travels at synchronous speed  appears stationary with respect to the rotor; and  has a sinusoidal space distribution  mmf wave due to one phase: Figure 3.7: Spatial mmf wave of phase a SM - 6 1539pk
  • 7.
    Balanced Steady StateOperation  The mmf wave due to the three phases are: MMFa = Kia cos γ ia = Im cos( ωs t )  2π   2π  MMFb = Kib cos γ −  ib = Im cos ωs t −   3   3   2π   2π  MMFc = Kic cos γ +  ia = lm cos ωs t +   3   3  MMFtotal = MMFa + MMFb + MMFc 3 = KIm cos( γ − ωs t ) 2 SM - 7 1539pk
  • 8.
    Balanced Steady StateOperation  Magnitude of stator mmf wave and its relative angular position with respect to rotor mmf wave depend on machine output  for generator action, rotor field leads stator field due to forward torque of prime mover;  for motor action rotor field lags stator field due to retarding torque of shaft load Figure 3.8: Stator and rotor mmf wave shapes SM - 8 1539pk
  • 9.
    Transient Operation  Stator and rotor fields may:  vary in magnitude with respect to time  have different speed  Currents flow not only in the field and stator windings, but also in:  damper windings (if present); and  solid rotor surface and slot walls of round rotor machines Figure 3.4: Current paths in a round rotor SM - 9 1539pk
  • 10.
    Direct and QuadratureAxes  The rotor has two axes of symmetry  For the purpose of describing synchronous machine characteristics, two axes are defined:  the direct (d) axis, centered magnetically in the centre of the north pole  The quadrature (q) axis, 90 electrical degrees ahead of the d-axis Figure 3.1: Schematic diagram of a 3-phase synchronous machine SM - 10 1539pk
  • 11.
    Mathematical Descriptions ofa Synchronous Machine  For purposes of analysis, the induced currents in the solid rotor and/or damper windings may be assumed to flow in two sets of closed circuits  one set whose flux is in line with the d-axis; and  the other set whose flux is along the q-axis  The following figure shows the circuits involved Figure 3.9: Stator and rotor circuits SM - 11 1539pk
  • 12.
    Review of MagneticCircuit Equations (Single Excited Circuit)  Consider the elementary circuit of Figure 3.10 dΨ ei = dt dΨ e1 = + ri dt Ψ = Li  The inductance, by definition, is equal to flux linkage per unit current φ L=N = N2P i where P = permeance of magnetic path  = flux = (mmf) P = NiP Figure 3.10: Single excited magnetic circuit SM - 12 1539pk
  • 13.
    Review of MagneticCircuit Equations (Coupled Circuits)  Consider the circuit shown in Figure 3.11 dΨ1 e1 = + r1i1 dt dΨ2 e2 = + r2i2 dt Ψ1 = L11i1 + L21i2 Ψ2 = L21i1 + L22i2 with L11 = self inductance of winding 1 L22 = self inductance of winding 2 L21 = mutual inductance between winding 1 and 2 Figure 3.11: Magnetically coupled circuit SM - 13 1539pk
  • 14.
    Basic Equations ofa Synchronous Machine  The equations are complicated by the fact that the inductances are functions of rotor position and hence vary with time  The self and mutual inductances of stator circuits vary with rotor position since the permeance to flux paths vary Iaa = L al + Igaa = L aa 0 + L aa 2 cos 2θ  2π  Iab = Iba = −L ab 0 + L ab 2 cos 2θ −   3   π = −L ab 0 − L ab 2 cos 2θ +   3  The mutual inductances between stator and rotor circuits vary due to relative motion between the windings Iafd = L afd cos θ Iakd = L akd cos θ  π Iakq = L akq cos θ +  = −L akq sin θ  2 SM - 14 1539pk
  • 15.
    Basic Equations ofa Synchronous Machine  Dynamics of a synchronous machine is given by the equations of the coupled stator and rotor circuits  Stator voltage and flux linkage equations for phase a (similar equations apply to phase b and phase c) dΨa ea = − R aia = pΨa − R aia dt Ψa = −laaia −labib − lacic + lafdifd + lakdikd + lakqikq  Rotor circuit voltage and flux linkage equations e fd = pΨfd + R fdifd 0 = pΨkd + R kdikd 0 = pΨkq + R kqikq ψ fd = L ffdifd + L fkdikd   2π   2π   − L afd ia cos θ + ib cos θ −  + ic cos θ +    3   3  ψ kd = L fkdifd + Lkkdikd   2π   2π   − L afd ia cos θ + ib cos θ −  + ic cos θ +    3   3  ψ kq = Lkkdikq   2π   2π   + L akq ia sin θ + ib sin θ −  + ic sin θ +    3   3  SM - 15 1539pk
  • 16.
    The dqo Transformation  The dqo transformation, also called Park's transformation, transforms stator phase quantities from the stationary abc reference frame to the dqo reference frame which rotates with the rotor   2π   2π    cos θ cos θ −  cos θ +    3   3  id  ia  i  = 2  2π   2π   ib    3 q  − sin θ − sin θ − 3  − sin θ + 3       i  i0   1 1 1  c    2  2 2  The above transformation also applies to stator flux linkages and voltages  With the stator quantities expressed in the dqo reference frame  all inductances are independent of rotor position (except for the effects of magnetic saturation)  under balanced steady state operation, the stator quantities appear as dc quantities  during electromechanical transient conditions, stator quantities vary slowly with frequencies in the range of 1.0 to 3.0 Hz The above simplify computation and analysis of results. SM - 16 1539pk
  • 17.
    Physical Interpretation ofdqo Transformation  The dqo transformation may be viewed as a means of referring the stator quantities to the rotor side  In effect, the stator circuits are represented by two fictitious armature windings which rotate at the same speed as the rotor; such that:  the axis of one winding coincides with the d-axis and that of the other winding with the q-axis  The currents id and iq flowing in these circuits result in the same mmf's on the d- and q-axis as do the actual phase currents  The mmf due to id and iq are stationary with respect to the rotor, and hence:  act on paths of constant permeance, resulting in constant self inductances (Ld, Lq) of stator windings  maintain fixed orientation with rotor circuits, resulting in constant mutual inductances SM - 17 1539pk
  • 18.
    Per Unit Representation  The per unit system is chosen so as to further simplify the model  The stator base quantities are chosen equal to the rated values  The rotor base quantities are chosen so that:  the mutual inductances between different circuits are reciprocal (e.g. Lafd = Lfda)  the mutual inductances between the rotor and stator circuits in each axis are equal (e.g., Lafd = Lakd) The P.U. system is referred to as the "Lad base reciprocal P.U. system"  One of the advantages of having a P.U. system with reciprocal mutual inductances is that it allows the use of equivalent circuits to represent the synchronous machine characteristics SM - 18 1539pk
  • 19.
    P.U. Machine Equationsin dqo reference frame  The equations are written with the following assumptions and notations:  t is time in radians  p = d/dt  positive direction of stator current is out of the machine  each axis has 2 rotor circuits  Stator voltage equations e d = pψ d − ψ qωr − R aid e q = pψ q + ψ dωr − R aiq e 0 = pψ 0 − R a i 0  Rotor voltage equations e fd = pψ fd + R fdifd 0 = pψ 1d + R 1di1d 0 = pψ 1q + R 1qi1q 0 = pψ 2q + R 2qi2q SM - 19 1539pk
  • 20.
    P.U. Machine Equationsin dqo Reference Frame (cont'd)  Stator flux linkage equations ψ d = −( Lad + Ll ) id + Lad ifd + Lad i1d ψ q = − ( Laq + Ll ) iq + Laq i1q + Laq i 2 q ψ 0 = −L0 i0  Rotor flux linkage equations ψ fd = L ffdifd + L f 1di1d − L adid ψ 1d = L f 1difd + L11di1d − L adid ψ 1q = L11qi1q + L aqi2q − L aqiq ψ 1q = L aqi1q +L 22qL 2 q − L aqiq  Air-gap torque T e = ψ diq − ψ qid SM - 20 1539pk
  • 21.
    Steady State AnalysisPhasor Representation For balanced, steady state operation, the stator voltages may be written as: e a = Em cos( ωt + α ) eb = Em cos( ωt − 2π 3 + α ) e c = Em cos( ωt + 2π 3 + α ) with ω = angular velocity = 2πf α = phase angle of ea at t=0 Applying the d,q transformation, e d = Em cos( ωt + α − θ ) e q = Em sin( ωt + α − θ ) At synchronous speed, the angle θ is given by θ = ωt + θ0 with θ = value of θ at t = 0 Substituting for θ in the expressions for ed and eq, e d = Em cos( α − θ 0 ) e q = Em sin( α − θ 0 ) SM - 21 1539pk
  • 22.
    Steady State AnalysisPhasor Representation (cont'd)  The components ed and eq are not a function of t because rotor speed ω is the same as the angular frequency ω of the stator voltage. Therefore, ed and eq are constant under steady state. In p.u. peak value Em is equal to the RMS value of terminal voltage Et. Hence, e d = E t cos( α −θ0 ) e q = E t sin( α −θ0 )  The above quantities can be represented as phasors with d-axis as real axis and q-axis as imaginary axis Denoting δi, as the angle by which q-axis leads E e d = E t sin δi e q = E t cos δi SM - 22 1539pk
  • 23.
    Steady State AnalysisPhasor Representation (cont'd)  The phasor terminal voltage is given by ~ in the d-q coordinates E t = ed + je q in the R-I coordinates = E R + jE l  This provides the link between d,q components in a reference frame rotating with the rotor and R, I components associated with the a.c. circuit theory  Under balanced, steady state conditions, the d,q,o transformation is equivalent to  the use of phasors for analyzing alternating quantities, varying sinusoidally with respect to time  The same transformation with θ = ωt applies to both  in the case of machines, ω = rotor speed  in the case of a.c. circuits, ω = angular frequency SM - 23 1539pk
  • 24.
    Internal Rotor Angle  Under steady state e d = −ωψq − idR a = ωL qiq − idR a = X qiq −idR a Similarly e q = ωψd − iqR a = − X did + X adifd − iqR a  Under no load, id=iq=0. Therefore, ψ q = −L qiq = 0 ψ d = L adifd ed = 0 e q = L adifd ~ and E t = e d + jeq = jL adifd  Under no load, Et has only the q-axis component and δi=0. As the machine is loaded, δi increases. Therefore, δi is referred to as the load angle or internal rotor angle.  It is the angle by which q-axis leads the phasor Et SM - 24 1539pk
  • 25.
    Electrical Transient Performance  To understand the nature of electrical transients, let us first consider the RL circuit shown in Figure 3.24 with e = Emsin (ωt+α). If switch "S" is closed at t=0, the current is given by di e=L + iR dt solving ( L) t Em sin( ωt + α − φ) −R i = Ke + Z  The first term is the dc component. The presence of the dc component ensures that the current does not change instantaneously. The dc component decays to zero with a time constant of L/R Figure 3.24: RL Circuit SM - 25 1539pk
  • 26.
    Short Circuit Currentsof a Synchronous Machine  If a bolted three-phase fault is suddenly applied to a synchronous machine, the three phase currents are shown in Figure 3.25. Figure 3.25: Three-phase short-circuit currents SM - 26 1539pk
  • 27.
    Short Circuit Currentsof a Synchronous Machine (cont'd)  In general, fault current has two distinct components: a) a fundamental frequency component which decays initially very rapidly (a few cycles) and then relatively slowly (several seconds) to a steady state value b) a dc component which decays exponentially in several cycles  This is similar to the short circuit current in the case of the simple RL circuit. However, the amplitude of the ac component is not constant  internal voltage, which is a function of rotor flux linkages, is not constant  the initial rapid decay is due to the decay of flux linking the subtransient circuits (high resistance)  the slowly decaying part of the ac component is due to the transient circuit (low resistance)  The dc components have different magnitudes in the three phases SM - 27 1539pk
  • 28.
    Elimination of dcComponent by Neglecting Stator Transients  For many classes of problems, considerable computational simplicity results if the effects of ac and dc components are treated separately  Consider the stator voltage equations e d = pψ d − ω ψ q − idR a e q = pψ q + ω ψ d − iqR a transformer voltage terms: pψd, pψq speed voltage terms: ω ψ q , ω ψ d  The transformer voltage terms represent stator transients:  stator flux linkages (ψd, ψq) cannot change instantaneously  result in dc offset in stator phasor current  If only fundamental frequency stator currents are of interest, stator transients (pψd, pψq) may be neglected. SM - 28 1539pk
  • 29.
    Short Circuit Currentswith Stator Transients Neglected  The resulting stator phase currents following a disturbance has the wave shape shown in Figure 3.27  The short circuit has only the ac component whose amplitude decays  Regions of subtransient, transient and steady state periods can be readily identified from the wave shape of phase current Figure 3.27: Fundamental frequency component of short circuit armature current SM - 29 1539pk
  • 30.
    Synchronous Machine Representationin System Stability Studies  Stator Transients (pψd, pψq) are usually neglected  accounts for only fundamental frequency components of stator quantities  dc offset either neglected or treated separately  allows the use of steady-state relationships for representing the transmission network  Another simplifying assumption normally made is setting ω = 1 in the stator voltage equations  counter balances the effect of neglecting stator transients so far as the low-frequency rotor oscillations are concerned  with this assumption, in per unit air-gap power is equal to air-gap torque (See section 5.1 of book for details) SM - 30 1539pk
  • 31.
    Equation of Motion(Swing Equation)  The combined inertia of the generator and prime- mover is accelerated by the accelerating torque: dωm J = Ta = Tm − Te dt where Tm = mechanical torque in N-M Te = electromagnetic torque in N-m J = combined moment of inertia of generator and turbine, kg•m2 m = angular velocity of the rotor in mech. rad/s t = time in seconds SM - 31 1539pk
  • 32.
    Equation of Motion(cont'd)  The above equation can be normalized in terms of per unit inertia constant H 1 Jω2 m H= 0 2 VA base where  0m = rated angular velocity of the rotor in mechanical radians per second  Equation of motion in per unit form is d ωr 2H = Tm − Te dt where ωm ωr = = per unit rotor angular velocity ω0m Tmω0m Tm = VA base = per unit mechanical torque Te ω0m Te = VAbase = per unit electromechanical torque  Often inertia constant M = 2H used SM - 32 1539pk
  • 33.
    Magnetic Saturation  Basic equations of synchronous machines developed so far ignored effects of saturation  analysis simple and manageable  rigorous treat a futile exercise  Practical approach must be based on semi- heuristic reasoning and judiciously chosen approximations  consideration to simplicity, data availability, and accuracy of results  Magnetic circuit data essential to treatment of saturation given by the open-circuit characteristic (OCC) SM - 33 1539pk
  • 34.
    Assumptions Normally Madein the Representation of Saturation  Leakage inductances are independent of saturation  Saturation under loaded conditions is the same as under no-load conditions  Leakage fluxes do not contribute to iron saturation  degree of saturation determined by the air-gap flux  For salient pole machines, there is no saturation in the q-axis  flux is largely in air  For round rotor machines, q-axis saturation assumed to be given by OCC  reluctance of magnetic path assumed homogeneous around rotor periphery SM - 34 1539pk
  • 35.
    The effects of saturation is represented as L ad = K sdL adu (3.182) L aq = K sqL aqu (3.183) Ladu and Laqu are unsaturated values. The saturation factors Ksd and Ksq identify the degrees of saturation.  As illustrated in Figure 3.29, the d-axis saturation is given by The OCC.  Referring to Figure 3.29, ΨI = Ψat 0 − Ψat (3.186) Ψat (3.187) K sd = Ψat + ΨI  For the nonlinear segment of OCC, can be ΨI expressed by a suitable mathematical function: ΨI = A sat eBsat ( Ψat − ΨTI ) (3.189) SM - 35 1539pk
  • 36.
    Open-Circuit Characteristic (OCC)  Under no load rated speed conditions id = iq = Ψq = e d = 0 E t = e q = Ψd = L adifd  Hence, OCC relating to terminal voltage and field current gives saturation characteristic of the d-axis Figure 3.29: Open-circuit characteristic showing effects of saturation SM - 36 1539pk
  • 37.
    For salient pole machines, since q-axis flux is largely in air, Laq does not vary significantly with saturation  Ksq=1 for all loading conditions  For round rotor machines, there is saturation in both axes  q-axis saturation characteristic not usually available  the general industry practice is to assume Ksq = Ksd  For a more accurate representation, it may be desirable to better account for q-axis saturation of round rotor machines  q-axis saturates appreciably more than the d- axis, due to the presence of rotor teeth in the magnetic path  Figure 3.32 shows the errors introduced by assuming q-axis saturation to be same as that of d-axis, based on actual measurements on a 500 MW unit at Lambton GS in Ontario  Figure shows differences between measured and computed values of rotor angle and field current  the error in rotor angle is as high as 10%, being higher in the underexcited region  the error in the field current is as high as 4%, being greater in the overexcited region SM - 37 1539pk
  • 38.
    The q-axis saturation characteristic is not readily available  It can, however, be fairly easily determined from steady-state measurements of field current and rotor angle at different values of terminal voltage, active and reactive power output  Such measurements also provide d-axis saturation characteristics under load  Figure 3.33 shows the d- and q-axis saturation characteristics derived from steady-state measurements on the 500 MW Lambton unit Figure 3.33: Lambton saturation curves derived from steady-state field current and rotor angle measurements SM - 38 1539pk
  • 39.
    Example 3.3  Considers the 555 MVA unit at Lambton GS and examines  the effect of representing q-axis saturation characteristic distinct from that of d-axis  the effect of reactive power output on rotor angle  Table E3.1 shows results with q-axis saturation assumed same as d-axis saturation Table E3.1 Pt Qt Ea (pu) Ksd δi (deg) ifd (pu) 0 0 1.0 0.889 0 0.678 0.4 0.2 1.033 0.868 25.3 1.016 0.9 0.436 1.076 0.835 39.1 1.565 0.9 0 1.012 0.882 54.6 1.206 0.9 -0.2 0.982 0.899 64.6 1.089  Table E3.2 shows results with distinct d- and q-axis saturation representation Table E3.2 Pt Qt Ksq Ksd δi (deg) ifd (pu) 0 0 0.667 0.889 0 0.678 0.4 0.2 0.648 0.868 21.0 1.013 0.9 0.436 0.623 0.835 34.6 1.559 0.9 0 0.660 0.882 47.5 1.194 0.9 -0.2 0.676 0.899 55.9 1.074 SM - 39 1539pk
  • 40.
    Simplified Models forSynchronous Machines  Neglect of Amortisseurs  first order of simplification  data often not readily available  Classical Model (transient performance)  constant field flux linkage  neglect transient saliency (x'd = x'q) Et x′d E´  Steady-state Model  constant field current  neglect saliency (xd = xq = xs) Et xs Eq = Xadifd Eq SM - 40 1539pk
  • 41.
    Reactive Capability Limitsof Synchronous Machines  In voltage stability and long-term stability studies, it is important to consider the reactive capability limits of synchronous machines  Synchronous generators are rated in terms of maximum MVA output at a specified voltage and power factor which can be carried continuously without overheating  The active power output is limited by the prime mover capability  The continuous reactive power output capability is limited by three considerations  armature current limit  field current limit  end region heating limit SM - 41 1539pk
  • 42.
    Armature Current Limit  Armature current results in power loss, and the resulting heat imposes a limit on the output The per unit complex output power is ~ * S = P + jQ = E t ~t = E t It ( cos φ + j sin φ ) I where Φis the power factor angle  In a P-Q plane the armature current limit, as shown in Fig. 5.12, appears as a circle with centre at the origin and radius equal to the MVA rating Fig 5.12: Armature current heating limit SM - 42 1539pk
  • 43.
    Field Current Limit  Because of the heating resulting from RfdI2fd power loss, the field current imposes the second limit  The phasor diagram relating Et, It and Eq (with Ra neglected) is shown in Fig. 5.13 Equating the components along and perpendicular to the phasor Et X adifd sin δ i = X slt cos φ X adifd cos δ i = E t + X slt sin φ Therefore X ad P = E tlt cos φ = E tifd sin δi Xs X E2 Q = E tlt sin φ = ad E tifd cos δi − t Xs Xs  The relationship between P and Q for a given field current is a circle centered at on the Q-axis and with as the radius. The effect of the maximum field current on the capability of the machine is shown in Fig. 5.14  In any balanced design, the thermal limits for the field and armature intersect at a point (A) which represents the machine name-plate MVA and power factor rating SM - 43 1539pk
  • 44.
    Field Current Limit Fig.5.13: Steady state phasor diagram Fig. 5.14: Field current heating limit SM - 44 1539pk
  • 45.
    End Region HeatingLimit  The localized heating in the end region of the armature affects the capability of the machine in the underexcited condition  The end-turn leakage flux, as shown in Fig. 5.15, enters and leaves in a direction perpendicular (axial) to the stator lamination. This causes eddy currents in the laminations resulting in localized heating in the end region  The high field currents corresponding to the overexcited condition keep the retaining ring saturated, so that end leakage flux is small. However, in the underexcited region the field current is low and the retaining ring is not saturated; this permits an increase in armature end leakage flux  Also, in the underexcited condition, the flux produced by the armature current adds to the flux produced by the field current. Therefore, the end-turn flux enhances the axial flux in the end region and the resulting heating effect may severely limit the generator output, particularly in the case of a round rotor machine  Fig. 5.16 shows the locus of end region heating limit on a P-Q plane SM - 45 1539pk
  • 46.
    End Region HeatingLimit Fig. 5.15: Sectional view end region of a generator Fig. 5.16: End region heating limit SM - 46 1539pk
  • 47.
    Reactive Capability Limitof a 400 MVA Hydrogen Cooled Steam Turbine Generator  Fig. 5.18 shows the reactive capability curves of a 400 MVA hydrogen cooled steam turbine driven generator at rated armature voltage  the effectiveness of cooling and hence the allowable machine loading depends on hydrogen pressure  for each pressure, the segment AB represents the field heating limit, the segment BC armature heating limit, and the segment CD the end region heating limit Fig. 5.18: Reactive capability curves of a hydrogen cooled generator at rated voltage SM - 47 1539pk
  • 48.
    Effect of Changesin Terminal Voltage Et Fig. 5.17: Effect of reducing the armature voltage on the generator capability curve SM - 48 1539pk