Synchronous machine device

SM1

SM
N

Synchronous machine device ........................................................................................ 1

1 Description................................................................................................................. 1
1.1 The synchronous machine symbol ................................................................ 2
1.2 Theoretical summary ....................................................................................... 2
1.3 Flux equations................................................................................................... 5
1.4 Voltage equations ............................................................................................ 5
1.5 Modeling saturation.......................................................................................... 6
1.6 Steady-state initialization ................................................................................ 8
1.7 Equations for mechanical part........................................................................ 9
1.8 Electrical parameters ..................................................................................... 10
2 Parameters and Rules ........................................................................................... 12
2.1 Main data tab .................................................................................................. 12
2.2 Electrical data tab ........................................................................................... 12
2.3 Mechanical data tab ....................................................................................... 13
2.4 Precision data tab .......................................................................................... 13
2.5 Control tab ....................................................................................................... 14
2.6 Observe/Scopes tab ...................................................................................... 14
2.7 Base quantities ............................................................................................... 14
2.8 Rules ................................................................................................................ 15
3 Netlist format ........................................................................................................... 16
4 Load-Flow model .................................................................................................... 19
5 Steady-state model ................................................................................................ 19
6 Initial conditions ...................................................................................................... 20
7 Frequency Scan model ......................................................................................... 20
8 Time-domain representation ................................................................................ 20
9 References .............................................................................................................. 22
Sébastien Dennetière, Jean Mahseredjian, 29/05/2012 9:44:26 PM

1 Description
This device is used to represent the usual design of a synchronous machine with 3-phase ac
armature windings on the stator. Up to 3 windings may be modeled on rotor d-axis (a field
winding and two dampers). Three more windings can be modeled on the rotor q-axis.
The model represents both generator and motor behavior.

EMTP-EMTPWorks, 6/19/2012 12:49:00 AM Page 1 of 22

1.1 The synchronous machine symbol
The synchronous machine device is a 3-phase device. Its symbol is automatically updated
according to internal winding configuration. The following figure presents available device pins.

Control bundle for

control signals

3-phase pin for

stator connections

SM1

SM
N
1-phase pin for stator
external neutral

Control bundle for

observable signals

Figure 1 The synchronous machine symbol

1.2 Theoretical summary

A detailed background to the machine modeling method can be found in [1] which is also referring
to previously available EMTP models. The model developed for this version of EMTP has several
improvements and new calculation methods.
The aim of this section is to provide some background notions that can help understanding the
modeling approach and the data requirements.
The following table presents a listing of the main variables used in the formulation of the machine
equations.

Symbol Description
v d , v q , v0 Stator voltages in dq0 domain
v f , vD1  vD2  vQ1  vQ2  vQ3  0 Field voltage v f . All other rotor voltages equal zero.
id , iq , i0 Stator currents in dq0 domain
iD1 , iQ1 , iD2 , iQ2 , iQ3 , if Rotor currents on d and q axis
p Number of poles
el Electrical speed (rad/s)
 2
Mechanical speed     el
p
b Base electrical speed (rad/s)
bm 2
Base mechanical speed bm    b
p

EMTP-EMTPWorks, 6/19/2012 12:49:00 AM Page 2 of 22

The Park’s transformation is initially used with saturation neglected so as to obtain linear relations
and apply superposition.
It is assumed that the resistance of each winding is constant.

The magnetic circuits of the rotor windings are assumed to be symmetric with respect to the
direct axis, which is the axis of the field created by the rotor. The quadrature axis is located 90
degrees behind the direct axis.

The self and mutual inductances of the armature windings are a constant plus a second harmonic
sinusoidal function of the rotor position. The mutual inductances between any winding of the field
structure and any armature winding is a sinusoidal function of the rotor position.

A current in any winding produces a magnetic field in the air gap with sinusoidal distribution. It
can be decomposed along the direct and quadrature axes. With this assumption, the effect of
harmonics in field distribution is negligible for a correctly dimensioned machine.

The hysteresis effects are neglected.

Eddy currents are negligible, except in the case of a solid rotor (cylindrical rotor) machine where
they can be represented by a q-axis winding.

In this version of EMTP, the synchronous machine consists of 9 windings :

 3 for the stator, one for each phase. They are denoted with the indices a, b, c,
respectively.
 6 for the rotor. On the axis of the rotor field, which is the direct axis, there are the
excitation winding f and a maximum of 2 damper windings denoted by D1 and D2. On the
quadrature axis, there are a maximum of 3 damper windings denoted Q1, Q2 and Q3.

The machine is represented by the diagram of Figure 2.

EMTP-EMTPWorks, 6/19/2012 12:49:00 AM Page 3 of 22

 Stator
b axis

D-axis

D2

D1

f 

Q1
Stator
c axis Q2
Stator
Q3 a axis

Q-axis

Figure 2 Schematic representation of the machine in DQ

The Park’s transformation matrix is defined by:
  2   2  
cos  cos    3  cos    3  0 0 0 0 0 0 
     
  2   2  
 sin  sin     sin     0 0 0 0 0 0 
  3   3  
 1 1 1 
 0 0 0 0 0 0 
 2 2 2 
 3 
 0 0 0 0 0 0 0 0 
 2 
2 
3
P 0 0 0 0 0 0 0 0 
3 2 
 
 0 3
0 0 0 0 0 0 0 
 2 
 
 0 0 0 0 0 0
3
0 0 
 2 
 
 3 
 0 0 0 0 0 0 0
2
0 
 
 3
 0 0 0 0 0 0 0 0 
 2  (1)
This transformation is power invariant. Physically, the three coils of the stator are replaced by
three windings 0, d, q revolving at the speed of the rotor and giving the same field and the same
induced voltage. The speed of the reference frame is:

EMTP-EMTPWorks, 6/19/2012 12:49:00 AM Page 4 of 22

 d
el 
dt (2)
The schematic representation of the machine is now given by the diagram of Figure 3.

Stator
b axis

D-axis

D2

D1

Q1

Q2

Stator Q3 Stator
c axis a axis
q

Q-axis

Figure 3 Schematic representation of the machine after Park’s transformation

1.3 Flux equations

The circuit consists of a maximum 9 coupled windings. By knowing the expression of each
winding inductance as a function of  , writing the relation between fluxes and currents in phase-
domain and applying the Park’s transformation of (1) results into:
  d   Ld 0 0 Mdf MdD1 MdD2 0 0 0  i 
   0 MqQ3   i 
d
L 0 0 0 0 M M
 q   q qQ1 qQ2
 q 
   0 0 L0 0 0 0 0 0 0  i 
 0    0 
 f   Mdf 0 0 L ff MfD1 MfD2 0 0 0  i 
f (3)
     MdD1 0 0 MfD1 LD1 MD1D2 0 0 0  i 
 D1     D1 
 D2  MdD2 0 0 MfD2 MD1D2 LD2 0 0 0  i 
   0  D2
MQ1Q3   i 
  Q1  
M 0 0 0 0 L M
qQ1 Q1 Q1Q2  Q1 
  Q2   0 MqQ2 0 0 0 0 MQ1Q2 LQ2 MQ2Q3  iQ2 
    
  Q3   0 MqQ3 0 0 0 0 MQ1Q3 MQ2Q3 LQ3  iQ3 
where Mij gives the mutual inductance between coils i and j.

1.4 Voltage equations

The rotor consists of 6 windings, the excitation winding f producing the flux in the machine, and 5
short-circuited damper windings D1, D2, Q1, Q2 and Q3.

EMTP-EMTPWorks, 6/19/2012 12:49:00 AM Page 5 of 22

For the stator, there are two possible setups: Wye (Y) or Delta-connection of the windings. Owing
to the symmetry of the stator circuit the resistances of the windings of the stator are equal and
constant:
ra  rb  rc  Ra (4)
Taking the Y connection circuit and establishing the Kirchoff’s law for each phase, results into a
relation that can be again rewritten through Park’s transformation:
 vd  Ra 0 0 0 0 0 0 0 0   id    d   el  q 
v  0 R  i     
 q  a 0 0 0 0 0 0 0  q   q   el  d 
v  0 0 Ra 0 0 0 0 0 0   i0     0 
 0     0   
 v f  0 0 0 rf 0 0 0 0 0   if   f   0  (5)
0   iD1    D1    0 
 0    0 d
0 0 0 rD1 0 0 0
      dt    
 0  0 0 0 0 0 rD2 0 0 0  iD2   D2   0 
  0 0 0 0 0 0 rQ1 0 0   iQ1    
  Q1   0 

 0   
 0  0 0 0 0 0 0 0 rQ2 0  iQ2    Q2   0 
        
 0   0 0 0 0 0 0 0 0 rQ3  iQ3    Q3   0 
where r gives the resistance of the winding.

1.5 Modeling saturation

To increase the accuracy of this model it is necessary to introduce saturation. The relation giving
the flux with respect to the currents is no longer linear; it is not represented by a single straight
line. It is difficult to model saturation very accurately. The data requirements can also become
problematic. The following approximations have been made in EMTP:
 EMTP decomposes each flux into a leakage flux and an air gap flux for each of the
windings. The leakage flux  li is a flux passing only through the relevant winding i. It
does not affect the other coils. Its path lies mainly in air. Thus, for the software this flux is
not saturable. On the other hand the air-gap flux mi is considered to cross all the
windings for each of the axes. EMTP does not consider any flux crossing two windings
without passing through the third (for saturation calculations only). Its path lies mainly in
iron. It will consequently be subject to saturation.
 The homopolar flux  0 is not saturable.
The following equation separates the leakage flux from the mutual flux:
  md 
  
 mq 
 0 
 
  nf  
  d    ld      md 
     d n 
 q   lq    n  
        D1   md 
 0   0    nd  
  f    lf   
          nD2  
 D1   lD1    n   md 
 D2    lD2    d  
      
 
 Q1   lQ1    Q1   n
  Q2    lQ2    nq  mq

      
  Q3    lQ3   nQ2  
   mq 
 nq  
 
 nQ3  
   mq 

 nq   (6)
Here ld is the leakage flux and md is the flux linking all d-axis windings (total flux minus
leakage flux).

EMTP-EMTPWorks, 6/19/2012 12:49:00 AM Page 6 of 22

There are two options for modeling saturation: independent and total saturation.

The choice of independent saturation is suitable in the case of a salient pole machine since the
magnetic structure differs between the direct and the quadrature axes. For each of the total air
gap fluxes md and mq , EMTP uses a simplified saturation curve, represented by several
straight line segments (see Figure 4). This model disregards the cross flux.

md

Mds2
kd2

Mds1

kd1

Mdu

iTof axisd

Figure 4 Saturation model for the d-axis

The current iTd represents the sum of the currents of the windings d, f, D1 and D2 referred back
to the excitation coil f. The current iTq represents the sum of the currents of the windings q, Q1 ,
Q2 and Q3 referred back to the coil q.

The unsaturated equation is taken from equation (3) after subtracting the leakage flux:
mdu  Mdu iTd  Ld  Lld  id  Mdf if  MdD1D1
i  MdD2iD2 (7)
where u means unsaturated.
For the saturated case the equation of a line segment in Figure 4 can be combined with the
current iTd relation taken from the above equation:
Mds i
mds  Mds i iTd  kd i  Ld  Lld  id  Mdf if  MdD1iD1  MdD2iD2   kd i (8)
Mdu 
It is not necessary to know the number of turns in each winding. Only their ratios need to be
known. EMTP calculates them using the assumption of the existence of a single common air gap
flux. For the first ratio, for example, it finds:
nf Mdf
 (9)
nd Ld  Lld
Equation (6) can be replaced into equation (5) to account for saturation. The same approach is
used for q-axis.

The total saturation model is suitable for solid-rotor machines (round rotor). In this case,
saturation is taken into account only with regard to the total air gap flux which is the result of the
vector sum of the common total fluxes for the axes d and q. A single saturation curve is sufficient

EMTP-EMTPWorks, 6/19/2012 12:49:00 AM Page 7 of 22

to characterize the phenomenon. In EMTP, it is represented by several straight line segments as
shown in Figure 4, only now the d subscript is replaced by the T subscript for total.

To simplify the problem, EMTP decomposes this case into 2 independent saturations for the
direct and the quadrature axes. This allows the expressions established in the first case to be
used again. Since the machine has smooth poles (solid rotor), it is assumed that the magnetic
structure is independent of the position of the rotor. In other words, for the 2 axes d and q, the
saturation curves are the same as that given by the total air gap flux. Thus, the slopes Mdu and
Mqu are the same and equal to a common slope Mu .
For the d-axis, the mutual flux can be written again as:
md  kd i  Mds i iTd (10)

If it is expressed as a function of the total mutual flux m  md

2
 mq
2
, it becomes:
m   Ms i iT
md  mdu  kT i Mdu iTd (11)
mu Mu iT
By comparing the above two relations:
iT
kd i  kT i d
iT
(12)
Mdsi  Msi
(13)
with:
iT  iT2d  iT2q (14)
Thus, it is possible to reuse the equations established in the case of two independent saturations.

1.6 Steady-state initialization

The steady-state positive sequence (hat) equations can be written from equations (5) and (3)
respectively:
v̂ d  Ra ˆid  el ˆq
(15)
v̂ q  Ra ˆiq  el ˆd
(16)
ˆ
v̂ f  rf if
(17)
ˆ d  Ldˆid  Mdf ˆif (18)
ˆ  L ˆi
q qq (19)
The currents in the damper windings are zero.

EMTP performs steady-state initialization of the machine by replacing the machine with 3 ideal
positive sequence voltage sources. The current flowing into the voltage sources is taken from the
network solution and the Fortescue transformation is applied to extract the sequence currents
flowing into the machine. The sequence currents are used to calculate the d and q axis currents
and to obtain all the steady-state phasors of the machine. Initialization for both positive and
negative sequence currents is performed. This is an approximation for the unbalanced case since
the machine impedance is not represented.
The electromagnetic torque is also initialized from the steady-state solution.

In this version of EMTP, the machine impedance matrix is not represented for the steady-state
initialization. It will be included when the 3-phase load-flow option becomes available.

EMTP-EMTPWorks, 6/19/2012 12:49:00 AM Page 8 of 22

The computations to account for saturation in the steady-state model are based on an iterative
process for finding the operating segment in the saturation characteristic in steady-state. The
iterations are only internal to the machine model since the terminal voltages and thus currents do
not change due to the previous assumptions for steady-state.

1.7 Equations for mechanical part

A single mass representation is usually adequate for hydro units, since the generator and the
turbine are close to each other on the same shaft. It is not correct for thermal units, specially
when subsynchronous resonance or similar problems related to torsional vibrations are being
studied.
The model used by the software is a mass-spring system. It gives good results compared to finite
element models in the bandwidth of EMTP applications. It makes it possible to model the
subsynchronous resonances with good accuracy, and since this model is linear, it gives relations
which are simple to implement. However it remains valid for small angles between two masses.

Unlimited number of masses can be specified to define the rotor shaft. Each major element is
considered to be a rigid mass connected to adjacent elements by mass less springs. An external
torque can be applied on each individual mass. Figure 5 shows a typical 6-mass model
representing a turbine/alternator assembly.

High pressure High pressure Low pressure Low pressure

turbine A turbine B turbine A turbine B Alternator Exciter

Tm1 Tm2 Tm3 Tm4 Tm5 Tm6

HSP12 DSM12 HSP23 DSM23 HSP34 DSM34 HSP45 DSM45 HSP56 DSM56
J1 J2 J3 J4 J5 J6

DSR1 DSD1 DSR2 DSD2 DSR3 DSD3 DSR 4 DSD4 DSR5 DSD5 DSR6 DSD6

Figure 5 Diagram of the turbine/alternator assembly of the mechanical part

The n-spring-connected rotating masses are then described by the rotational form of Newton’s
second law:
dωrm
J  Dωrm  HSPθrm  Tturbine  Tgen/exc (20)
dt
where:
J is the diagonal matrix of moments of inertia ( J1 to J6 in Figure 5)
ωrm is the vector of mechanical speeds
θrm is the vector of angular positions
HSP is the tridiagonal matrix of stiffness coefficients
D is the tridiagonal matrix of damping coefficients
Tturbine is the vector of torques applied to the turbine stages (see Tm1 to Tm4 in Figure 5)
Tgen/exc is the vector of electromagnetic torques of generator and exciter (see Tm5 and Tm6 in
Figure 5)
The moment of inertia and the stiffness coefficients are normally available from design data. The
spring action of the shaft section between masses i-1 and i creates a torque which is proportional
to the angle twist rm i  rm i 1 . The proportionality factor is the stiffness coefficient or spring
constant HSPi i1 . This spring action torque acts in opposite directions on masses i-1 and i :
Tspring i 1  Tspring i  HSPi 1 i  rm i 1  rm i  (21)

EMTP-EMTPWorks, 6/19/2012 12:49:00 AM Page 9 of 22

From equation (21) it can be seen that the HSP matrix has the following form (3 masses case):
 HSP12 HSP12 0 
HSP   HSP12 HSP12  HSP23 HSP23  (22)
 0 HSP23 HSP23 
Three damping effects are included with the damping coefficients, namely the absolute speed self
damping DSDi of mass i, the mutual damping coefficient DSMi i 1 and the mutual damping
coefficient DSMi 1 i . The damping torque acting on mass i is therefore:

  
Tdamping i  DSDi rm i  DSMi 1 i rm i  rm i 1  DSMi i 1 rm i  rm i 1  (23)
From equation (23) it can be seen that D has the same structure as HSP except that the
diagonal element is now DSDi  DSMi 1 i  DSMi i 1 , as shown in this 3 masses example:
DSD1  DSM12 DSM12 0 

D   DSM12 DSD2  DSM12  DSM23 DSM23  (24)

 0 DSM23 DSD3  DSM23 
Another damping coefficient, named speed deviation self damping for a given mass (DSR), is
also available. By definition:
Ti  DSRi  m i  bm  (25)
where bm is the base mechanical (synchronous) speed.
It is very difficult to obtain realistic values for these damping coefficients. Fortunately, they have
very little influence on the peak torque value during transient disturbances.
The torques are known through the following relations:
Pturbine i
Tturbine i 
rm i
(26)

p
Tgenerator   diq   qid
2

(27)
The torque on the exciter is
v f if
Texciter 
exciter
(28)
where exciter is the mechanical speed of the exciter.

1.8 Electrical parameters

Before starting the simulation, it is necessary to determine the parameters used in the equations
of the machine. These may be given by the constructor. In this case, the user provides data in pu
and EMTP will internally convert them into SI units.
There is another method of characterizing the machine. It involves measurements obtained via
three-phase short-circuit tests or via frequency response tests defined by IEEE or IEC standards.
In this second case, EMTP must employ a parameter fitting method to calculate the self
inductances, mutual inductances and resistances of the machine.
The details of the calculations will be made available in a future release of EMTP. The material
that was used in coding the data conversion was based on the references [2]-[9]
It is noticed that this version has more options for data input and allows entering an extra damper
winding on each axis. The calculated machine data is available for each simulation in a separate
HTML file for machines.

It is recalled that according to IEEE and IEC the known quantities are:
o Ra : armature resistance
o Xl : armature leakage reactance

EMTP-EMTPWorks, 6/19/2012 12:49:00 AM Page 10 of 22

 o X0 : zero sequence reactance
o X'd , X'q : transient reactances
o X''d , X''q : subtransient reactances
o X'''d , X'''q : sub-subtransient reactances
o Td' , Tq' : transient short-circuit time constants
o Td'' , Tq'' : subtransient short-circuit time constants
o Td''' , Tq''' : sub-subtransient short-circuit time constants
' '
o Td0 , Tq0 : transient time constants in open-circuit
'' ''
o Td0 , Tq0 : sub-transient time constants in open-circuit
''' '''
o Td0 , Tq0 : sub-subtransient time constants in open-circuit

The Canay reactance Xc ( Lc ) is used to model the machine more accurately along the direct
axis. It takes into account a linkage air gap flux for the windings f and D1 in addition to the real
air gap flux common to the three windings d, f and D1. Figure 6 shows the equivalent with the
extra damper winding D2 circuit.

id La Ll1 Ll2 rfm L fm

+ + + +
+ +
+

rD1m rD2m
+

pd v fm
Md
LD1m LD2m

- -

Figure 6 Equivalent network on d-axis solved in EMTP

This circuit is constructed after rescaling the field structure quantities. The subscript m stands for
modified. It is noticed that:
Md  Ld  La (29)
The inductance La is the armature leakage reactance.
The damper winding circuit inductances must also account for Md in addition to rescaling. It can
be shown that:
1 1 1
  (30)
Lc  La Md Ll1
Another option is available through the G(p) (p is the differential operator in Laplace domain)
function. It allows defining extra time-constants and calculating Ll1 and Ll2 . It can be given by:
 d  p  Md 1  T1p 
G(p)   (31)
v f  p  
rfm 1  Td0
'

p 1  Td0
''
p 
for the presence of only one damper and by:

EMTP-EMTPWorks, 6/19/2012 12:49:00 AM Page 11 of 22

  d  p  Md 1  T1p 1  T2p 
G(p)   (32)
v f  p   
rfm 1  Td0p 1  Td0
' ''

p 1  Td0
'''
p 
for the presence of two damper windings on d-axis.

2 Parameters and Rules

2.1 Main data tab
The first tab “Main Data” allows selecting initialization options, entering the general data of the
machine and the magnetization characteristics if needed.

 Steady-state voltage and angle  : represents steady-state phasors at the terminals of

the machine. This data is generally available from a load-flow solution with given
constraints on the machine and PQ loads. In this version it can be taken from a separate
load-flow program. The next EMTP version will provide a load-flow feature that allows to
fix these values automatically.
 Load-Flow solution device: see Load-Flow section below.
 Frequency (f) is the steady-state operating (synchronous) frequency of the machine.
 Number of poles is the number of poles (p) which characterizes the machine rotor.
 Armature winding connection type can be Wye (Y), Wye grounded (internal ground) or
Delta.
 Rated line-to-line voltage is the Rated line-to-line stator terminal voltage used in the
definition of base values upon which pu machine parameters are assumed to apply.
 Rated power is the total 3-phase volt-ampere rating of the machine. Used in the
definition of base values upon which pu machine parameters are assumed to apply.
 Field current for 1 pu armature voltage is the value of field current which will produce
rated armature (1 pu) voltage on the air gap line. This quantity is also called i agline .
 Several machines on the same bus: if this option is chosen the user has to define the P
and Q power splits: proportionality factors which are used only to split the power among
the machines during initialization. If, for example, this is a dual machine, if PA is specified
PA
for this half of the dual, and PB for the other half, then the fraction will be
PA  PB
assigned to this half of the dual machine initialization.
 Simulate saturation allows entering the piecewise characteristic of the magnetization
curve of the machine. It can be determined from the no-load test curve of the machine.
Only the positive part is entered. The origin is assumed if not explicitly entered.
o The common saturation effect can be taken into account with the “Total
saturation” option. “Separate Saturation” simulates saturation on d-axis
separately from q-axis. The modeling methods are presented in the theoretical
section above.

2.2 Electrical data tab

The second data tab allows entering the data for making the electrical circuit of the machine on
each axis. Several data input options are available and tooltips provide identification for all
variables. Some details on the data conversion methods are provided in the above theoretical
section of this document.
It is noticed that for the q-axis parameters, when winding parameters are unavailable, the user
can choose the option “Undefined”. In this case XaQ1 = XaQ2 = XaQ3 =0, XQ1 = XQ2 = XQ3 =0 and
RQ1  RQ2  RQ3   for all types of q-axis data.
For each simulation involving synchronous machines, EMTP produces an HTML page showing
the calculated parameters for each machine. The page is available from the main Output HTML
file from the menu “EMTP>View Output Files”.

EMTP-EMTPWorks, 6/19/2012 12:49:00 AM Page 12 of 22

2.3 Mechanical data tab
 Share mechanical shaft option can be chosen if the mechanical shaft of the machine is
shared with another machine in the current circuit (or subcircuit). When this option is
selected the Mass Data of this machine is automatically made equal to the Mass Data of
the shared machine.
 Number of masses is used to enter the total number of masses to model the mechanical
shaft.
 Index of rotor mass: allows specifying the mass number of the generator rotor within the
interconnected mass-spring shaft system. This number must be positive and lower or
equal to the number of masses.
 Index of exciter mass allows specifying the mass number of the exciter mass on the
shaft system. If no exciter mass exists, use 0.
 Mass data: mechanical parameters for the shaft system gives the parameters that
describe the masses and connections between masses. See the above section
“2.3 Equations for mechanical part” for more details on the damping coefficients. It is
allowed to specify the fraction (%) of the total external mechanical torque which is
associated with the given mass. If the total is greater than 100% a scaling factor will be
applied.
 Use inertia constant H in seconds (s) instead of the Moment of inertia allows using
H instead of the Moment of inertia. This constant is defined as the ratio of the kinetic
energy of the rotating mass at base mechanical speed to the rated power:
J2
H  bm
2Sb
The user can click on the hyperlink title of this data grid to get more information on
entered data.
 Mass data: Observe, Scope and Control selections allows selecting the mechanical
variables that need to be observed by a control system, or need to be available under the
machine scopes, or need to be controlled by an external control system. The user can
click on the hyperlink title of this data grid to get more information on entered data. The
selected scopes become available under the machine scopes in the plot processing
package. Observe and control selections become available in the Observe and Control
bundles respectively.
 The mechanical angle of each mass (in rad) can be selected for observe: the stator
phase-a is the origin of the angles. The difference between the mechanical angle and the
synchronous angle of each mass (in deg) can be selected for scope.
 The mechanical speed (in rad/s or in pu) of each mass can be selected for scope and
observe.
 The torque (in MNm) between the different sections of the shaft can be selected for
scope and observe. The shaft torque Tm_i is the torque on the shaft section connecting
masses i and i+1.
 The steady-state mechanical power applied on mass i can only be observed.
 The external mechanical power applied on mass i can be controlled. If this control option
is chosen, the total power or total torque control can not be used.
 Use “pu” for Scope variables units: allows obtaining all the scope signals in pu.
 Use “pu” for Observe and Control variables units: allows obtaining and calculating all
the observe and control signals in pu.

2.4 Precision data tab

 Damping factor is the ratio between built-in damping resistors and discretized inductive
elements used in the machine model. Reduce this number when numerical stability
problems are encountered. The selected default data is suitable for most cases.
 Rotor speed tolerance: Relative tolerance associated with the iterative solution of rotor
speed at each solution time-point.

EMTP-EMTPWorks, 6/19/2012 12:49:00 AM Page 13 of 22

 Maximum number of iterations in rotor speed computation is the maximum number
of iterations allowed in the calculation of rotor speed at each solution time-point.
 Apply maximum precision: when this option is checked, an iterative procedure is
applied to achieve full convergence of machine equations with network equations.
Selecting this option provides an increased precision, but reduced computational speed.
 Relative tolerance: Relative tolerance associated with the iterative solution of voltage (if
the maximum precision option is checked) at the machine terminals at each solution time-
point.
More details are available in Figure 7.

2.5 Control tab

 Field voltage (on d-axis): if this option is selected, the field voltage on d-axis E f is
controlled in V or pu. See the section 2.7 for more explanations on this variable.
 Field voltage: if this voltage is selected then the controlled voltage is v f in V or pu (see
section 2.7). The symbol Vf is used in this tab instead of v f .
 Total mechanical torque: if this option is chosen the total torque (in MNm or pu) applied
on the mechanical shaft is controlled. If this option is selected the mechanical power
control of the previous data tab on any mass is no more available. Positive torque is used
for generator operation and negative torque is used for motor operation.
 Total mechanical power: If this option is chosen the total power (in MW or pu) applied
on the mechanical shaft is controlled. If this option is selected the mechanical power
control of the previous tab on any mass is no more available.
 Use “pu” for Control variables units: all control signals are entered in pu.

2.6 Observe/Scopes tab

The selections on this tab are self-explanatory.

 If the Observe option is selected, the corresponding signal becomes automatically

available in the observe signal bundle of the machine.
 If the Scope option is selected, the selected variable will become available under
machine scopes. It will be identified by the corresponding signal name followed by the
machine name.
Example:
id_SMX
is the signal name that becomes available for the machine named SMX and the variable
id .
 Use “pu” for Scope variables units: allows obtaining all the scope signals in pu.
 Use “pu” for Observe variables units: allows obtaining all the observe signals in pu.

2.7 Base quantities

The base quantities with rms value of a p-pole, three phase machine with rated line-to-line rms
voltage Vrated , rated value of angular frequency b ( bm for mechanical), and rated volt-ampere
Srated , are as follows (all entered data is first converted to fundamental units, no multiplicative
factors):

 Base rotor and stator voltages, rms line-to-line: Vb  Vrated

 Base volt-ampere Sb  Srated

EMTP-EMTPWorks, 6/19/2012 12:49:00 AM Page 14 of 22

 Sb
 RMS base stator current ibs 
3Vb
 Base rotor current ibr  Iagline
Vb2
 Base stator impedance Zbs 
Sb
pS
 Base torque Tb    b
 2  b
V
 Base flux b  b
b
The voltage rating is line-to-line for wye-connections. For delta-connections it is necessary to
multiply this voltage by 3 to assure that the base impedance Zbs is 3 times larger than in the
wye-connection.

It is noticed that when the machine is operating in open-circuit mode, equations (15) to (19) can
be used to write:
vq
Mdf  (33)
elif
which is also equivalent to:
Vb
Mdf  (34)
eliagline
The field voltage E f is calculated as follows:
vf
Ef  elMdf if  elMdf (35)
rf
The quantities E f and v f are identical when pu values are used assuming nominal frequency
conditions.

2.8 Rules
A number of tests in the internal EMTP code are designed to signal data errors and solution
problems. It is however impossible to intercept all errors and some conditions may conduct to
catastrophic error messages. The data forms can not intercept errors when undetermined (named
parameters) values are entered in data fields.
Since the symbol of this device is automatically redrawn through scripts, it is not allowed to
modify it through the symbol editor. It is not allowed to delete any pins or to modify pin attributes.

For the synchronous machine to initialize and operate properly it is mandatory to have steady-
state conditions established by other energy source types.

EMTP-EMTPWorks, 6/19/2012 12:49:00 AM Page 15 of 22

3 Netlist format
This device allows method-based scripting. The object data and methods are described in the
script file referenced by the device Script.Open.Dev attribute.
Example of data:

_SM;SM1a;4;2;s114a,s116,
?m,10,
_SM;SM1b;4;2;s114b,s116,
_SM;SM1c;4;2;s114c,s116,
V=22 22 22,1kVRMSLL,
P=0 -120 120,1,
f=60 2 1,
R=22 600 1200,
S=0 1 1 1 1 0 2 0,
1322 1.0
2200 1.2
;
Ra=.0045 .14 0.12
ds=0 2 1.65
Xdp=0.25 0.20 0.2273
Td0p=4.5 0.04 0.00215
Noex=1 0 0.0019 0.000075
Xc=0
qs=0 2 1.59
Xqp=0.46 0.20 0.244
Tq0p=0.55 0.09 0.0085
MS=0 4 2 1,
0 58.27 46.51 0 5.953E06 0
0 7423.5 5926.96 0 132.85E06 0
0.5 13091.06 10451.96 0 67.96E06 0
0.5 2102.8 1678.89 0 0 0
0 2 2 0 0
0 2 2 0 0
0 2 2 0 0
0 2 0 0 0
MSu=0 1 0,
PR=100 1e-06 300 0 1e-06,
CTRL=0 0 0 0,
OBSu=0 0,
OBS=0 0 0 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 …,

Field Description
_SM Part name
SM1a Instance name for the phase a stator coil, any name.
4 Total number of pins
2 Number of pins given in this data section
S114a Signal name connected to k-pin of the stator phase a coil, any
name
S116 Signal name connected to m-pin (Neutral) of the stator phase a coil,
any name
?m Request for scopes, sent to scope group Machine, optional
10 Number of scope outputs

EMTP-EMTPWorks, 6/19/2012 12:49:00 AM Page 16 of 22

SM Part name
SM1b Instance name for the phase b stator coil, any name.
4 Total number of pins
2 Number of pins given in this data section
S114b Signal name connected to k-pin of the stator phase b coil, any
name
S116 Signal name connected to m-pin (Neutral) of the stator phase b coil,
any name
ASM Part name
ASM1c Instance name for the phase c stator coil, any name.
4 Total number of pins
2 Number of pins given in this data section
S114c Signal name connected to k-pin of the stator phase c coil, any
name
S116 Signal name connected to m-pin (Neutral) of the stator phase c coil,
any name
V= Start steady state values section
22 Steady state voltage phase a
22 Steady state voltage phase b
22 Steady state voltage phase c
1kVRMSLL Units of steady state voltage: V, kV, MV, kVRMS, kVRRMSLL
P= Start Steady state phasor angle section
0 Steady state phasor angle at the terminals of the machine, phase a
-120 Steady state phasor angle at the terminals of the machine, phase b
120 Steady state phasor angle at the terminals of the machine, phase c
1 Units for steady state angle: degrees (1) or radians (1rad)
f= Start steady state operating frequency section and dummy
parameters
60 Steady state operating frequency
2 Number of poles
1 Armature winding connections: 0 for Wye grounded, 1 for Wye, 2
for Delta.
R= Start rated values section
22 Rated line to line voltage (kVRMS)
600 Rated power (MVA)
1200 Field current for 1 pu armature voltage
S= Start saturation values section
0 Several machines on the same bus
1 P power split
1 Q power split
1 Simulate saturation (0 for no saturation, 1 for saturation)
1 Total saturation (0 for no total saturation, 1 for total saturation)
0 Independent saturation (0 for no independent saturation, 1 for
independent saturation)
2 Number of points in the d-axis characteristic
2 Number of points in the q-axis characteristic
1322 1.0 Magnetization characteristic on d-axis (or both axis)
2200 1.2
Ra= Start stator windings parameters section
.0045 Armature resistance
.14 Armature leakage reactance
.12 Zero-sequence reactance
ds= Start d-axis data type section
0 0 for Open Circuit or 1 for Short Circuit or 2 for FREQ or 3 for PU

EMTP-EMTPWorks, 6/19/2012 12:49:00 AM Page 17 of 22

1 Number of dampers on the rotor
1.65 Self reactance for the d-axis of the armature
Xdp= Start reactance specifications section
0.25 d-axis transient reactance
0.20 d-axis subtransient reactance
0.2273 d-axis sub-subtransient reactance
Td0p= Start time constant specifications section
4.5 d-axis open-circuit transient time constant
0.04 d-axis open-circuit subtransient time constant
0.00215 d-axis open-circuit sub-subtransient time constant
Noex= Start extra data section
1 No extra data (1) or extra data (0)
0.0019 First constant for G(p) definition
0.000075 Second constant for G(p) definition
Xc=0 Canay reactance
2 Rotor type : 1 for single cage, 2 for wound, 3 for double cage.
qs= Start q-axis data type section
0 0 for Open Circuit or 1 for Short Circuit or 2 for FREQ or 3 for PU
2 Number of dampers on the rotor q-axis
1.59 Self reactance for the q-axis of the armature
Xdq= Start reactances specifications section
0.46 q-axis transient reactance
0.20 q-axis subtransient reactance
0.244 q-axis sub-subtransient reactance
Tq0p= Start time constant specifications section
0.55 q-axis open-circuit transient time constant
0.09 q-axis open-circuit subtransient time constant
0.0085 q-axis open-circuit sub-subtransient time constant
Ms= Start mechanical parameters section
0 Share mechanical; shaft (1 for yes)
4 Number of masses
2 Index of rotor mass
1 Index of exciter mass
Table of data Mechanical parameters table (4 masses in this example)
0220 Observable, scope and control table
0220
0220
0220
MSu= Start mechanical parameters units specification section
0 Use H instead of moment of inertia
1 Use “pu” for mechanical scope variable units
0 Use “pu” for mechanical observe and control variable units
PR= Start precision control data section
100 Damping factor
1e-6 Rotor speed tolerance
300 Maximum number of iteration in rotor speed computation
0 Apply maximum of precision when 1
1e-6 Voltage convergence tolerance
CTRL= Start Control section, 0 means 0ff and 1 means On
0 Control on field voltage (d-axis)
0 Control on external torque
0 Control on external power
0 Control on field voltage
OBSu= Start Scope and Observe in “pu” section

EMTP-EMTPWorks, 6/19/2012 12:49:00 AM Page 18 of 22

0 Use “pu” for scope variable units
1 Use “pu” for observe variable units
OBS= Start Scope and Observe requests
2 1 2 0 2 0 1 0……. Scope and Observe requests: 0 for none, 1 for scope request, 2 for
observe request, 3 for scope and observe request.

4 Load-Flow model
This machine does not participate (open-circuit) into the Load-Flow solution directly, but it can
provide its Thevenin circuit equivalent to the requesting Load-Flow (LF) solution device. It can
also reference a Load-Flow solution device (see “Load-Flow solution device” on the first data tab)
for automatically retrieving its steady-state voltage phasors from the Load-Flow solution file when
acting in the Steady-State solution.
The “Load-Flow solution device” can be an LF device or “BUS:”. It establishes a link (a reference)
for retrieving data from a load-flow solution. In this example (below) the machine SM1 is
disconnected in the load-flow solution and is referring to LF3 for its load-flow solution data
automatically retrieved by EMTP from the “Load-Flow solution data file” during the subsequent
Steady-state solution when “Start from Load-Flow solution” is turned on in the
EMTP>Simulation Options. Note also that LF3 is referring back to SM1 for obtaining the Thevenin
impedance of LF3 in the Load-Flow solution.

LF3
P=-31MW ?m
Q=16.7MVAR
LF

SM:SM1 12kV
SM

40MVA
PQbus:LF3
Pm
SM1

f(u) 1 Omega_1

If the selection is "BUS:" the connected stator signal name (bus) will be used to retrieve data. You
must enforce the signal name by making it visible. This method is optional; it is suggested to use
the LF device naming approach shown above.

A load function (“Load now” button) is also available on the first data tab to manually load the
steady-state voltage phasors and optionally eliminate the reference to the Load-Flow solution
device in the Steady-state initialization process.

5 Steady-state model
A steady-state solution is always computed by EMTP at the fundamental frequency of each
synchronous machine. At fundamental frequency the machine is replaced by 3 voltage sources.
The voltage of this source is specified by the user. All the machine parameters are initialized at
this step.
At non fundamental frequency steady-state solution the machine is represented by a harmonic
impedance matrix using techniques similar to [10]. The steady-state solution obtained with this
impedance is superimposed with the steady-state solution at fundamental frequency. A non
fundamental frequency steady-state solution can occur in some cases when EMTP must perform
a harmonic steady-state solution.

EMTP-EMTPWorks, 6/19/2012 12:49:00 AM Page 19 of 22

6 Initial conditions
Initial conditions are calculated from the steady-state solution as explained in “1.6Steady-state
initialization”. The synchronous machine can not start from no initial conditions, it needs a steady-
state solution before starting the time-domain solution.

7 Frequency Scan model

For a frequency scan study the synchronous machine is modeled by a harmonic impedance
matrix as in the steady-state case. No ideal voltage source is inserted during this calculation,
even for the fundamental frequency of the machine.

8 Time-domain representation
The time-domain representation of the synchronous machine has already been explained in the
above sections. In the time-domain solution this device is a nonlinear function. If the option “Apply
maximum precision” is chosen, it is solved through the iterative procedure of EMTP with all
nonlinear devices until convergence according to the relative tolerance option for each machine.
EMTP uses an internal speed loop for each synchronous machine.
Two parameters controlling the speed loop and the voltage loop can be modified by the user. The
Figure 7 shows the iterative process for solving the synchronous machine model.

EMTP-EMTPWorks, 6/19/2012 12:49:00 AM Page 20 of 22

Figure 7 The iterative solution process: speed and voltage loops

EMTP-EMTPWorks, 6/19/2012 12:49:00 AM Page 21 of 22

9 References
[1] H. Dommel: “EMTP Theory Book”, April 1996, Microtran Power System Analysis
Corporation.
[2] “IEEE Standard test procedures for synchronous machines”, IEEE Std 115-1995
[3] IEC, “Recommendations for Rotating Electric Machinery”, Publ. 34-4A, 1972
[4] “IEEE guide for synchronous generator modeling practices in stability analyses”, IEEE
Standard P1110, Chapter 6, version 2002
[5] S.D. Umans, J.A. Mallick, G.L. Wilson : “MODELING OF SOLID ROTOR
TURBOGENERATORS PART I and II”, IEEE transactions on Power Apparatus and
Systems, Vol. PAS-97, No. 1, January/February 1978
[6] Canay : “Modelling of alternating-current machines having multiple rotor circuits” IEEE
Transactions on energy conversion, Vol 8, No. 2, June 1993
[7] Canay : “Determination of the model parameters of machines from reactance operators
xd(p), xq(p) (evaluation of standstill frequency response test)”, IEEE Transactions on
energy conversion, Vol 8, No. 2, June 1993
[8] S.D. Umans, “Modelling of Solid Rotor Turbogenerator.” Report #46, Electric Power
Systems Engineering Laboratory, MIT, Cambridge, Mass., January 1976
[9] “Determination of synchronous machine stability study constants”, Volume 2, EPRI report
EL-1424.
[10] W. Xu, H. Dommel and J. Marti: “A synchronous machine model for three-phase harmonic
analysis and EMTP initialization”, IEEE Transactions on Power Ssytems, Vol. 6, No. 4,
November 1991, pages 1530-1538

EMTP-EMTPWorks, 6/19/2012 12:49:00 AM Page 22 of 22