Note: Descriptions are shown in the official language in which they were submitted.
20-TR-1183
ELECTRICAL PROPULSION PROCESS FOR A TRACTION
VEHICLE WITH AN ON-BOARD SOURCE OF POWER
~ackground of the Invention
This invention relates generally to a new and
improved process of propelling a vehicle, and it relates
more particularly to a propulsion method for an internally
powered traction vehicle that is driven by the combination
of a prime mover, electric power generating means, and
adjustable speed a-c electric motors.
In a large self-propelled electrically driven
traction vehicle, such as a locomotive or an off-highway
truck, the wheels of the vehicle are propelled (or
10 retarded) by electric motors energized by electric power -
generated by a rotating electrodynamic machine that in
turn is driven by an on-board thermal prime mover such as
a diesel engine. See for example U.S. patent No. 3,878,400 -
McSparran. Heretofore traction motors commonly have been
15 of the direct current ~d-c) variety, and the necessary d-c
excitation for such motors has been supplied either by
using a d-c generator or, alternatively, by rectifying
the output of an alternating current (a-c) generator. O
Because a-c motors are generally more simple
and compact than d-c motors, are not limited by commutator
constraints, are capable of greater tractive effort than
d-c motors at high speeds, and are relatively light
weight, low cost, and eas~ to maintain, persons working
in this art have been giving increasing attention to
utilizing adjustable speed a-c traction motors rather than
d-c motors in drive systems for eleetrically propelled
-1--
20-TR-1183
traction vehicles. In an a-c propulsion system, traction
motor speed (and hence vehicle speed) is dependent, in
large measure, on the fundamental frequency of the
excitation supplied to the stator windings of the motors,
and in order to control the frequency it has been proposed
to supply excitation to the motors via variable-frequency
electric power static inverters or frequency changers
formed by a plurality of controllable electric valves or
semiconductor switching elements (e.g., thyristors) of
the kind having the ability to hold off forward voltage
until turned "on" in response to a suitable firing or
gate signal. Once a valve is triggered or fired by its gate
signal, it switches from a blocking or non-conducting state
to a forward conducting state in which it can freely
conduct load current until this current is subsequently
transferred or commutated to a companion valve in the
power conversion apparatus. In the case of an inverter,
the valves are arranged in alternative paths of load-
current conduction between a set of a-c terminals (which are
connected to the stator windings of the a-c motor) and a pair
of d-c terminals (which are adpated to be coupled to a
suitable d-c power supply), and they are cyclically fired
in a predetermined sequence so as to convert d-c power
into a-c power having a fundamental frequency determined
by the switching frequency of the valves. Either voltage
or current source inverters can be utilized.
With a voltage source inverter, the amplitude
and frequency of the fundamental alternating voltage that
is supplied to the stator terminals of the associated a-c
motors are controlled, and stator current can vary widely
20-TR-1183
~S39~:
in magnitude. With a current source inverter (hereinafter
referred to as a controlled current inverter) the
quant ties that are controlled are the ampli~ude and
frequency of alternating current exciting the statox windings,
and the voltage magnitude can change rapidly during
commutation. A controlled current type of inverter is
required in practicing the present invention. For a
particular example of a circuit well suited for this
purpose, see the improved auto-sequential commutated
inverter that is described and claimed in U.S. patent
No. 3,980,941 - ~riebel.
For proper operation of a controlled current
inverter, its d-c terminals need to be coupled to a source
of rela~ively smooth direct current the magnitude of which
determines the amplitude of alternating current supplied
to the a-c traction motor, and this source needs to be
suitably controlled or regulated so as to set and maintain
a desired magnitude of direct current. In an internally
powered vehicle having an on-board prime mover, the d-c
source would logically comprise a rotating d-c generator
having its armature connected to the d-c terminals of e~ch
controlled current inverter by way of a d-c link including
a current smoothing filter, and current magnitude would be
regulated by appropriately controlling the electromagnetic
excitation of the generator to thereby adjust the
magnitude of voltage that the generator impresses on the
d-c link. The current smoothing filter ordinarily is in
the form of a series inductor (also referred to as a reactor
or choke), and its functions are to absorb short duration
voltage transients and to limit the rate of change of
_ 20-TR-1183
ll~S3g2
direct current that the generator supplies to each
separate inverter-motor set. More specifically, the
inductance provided by the filter in the d-c link, taken
together with any other source inductance and with the
motor inductance, prevents excessive current peaks that
might overstress the inverter valves, and it reduces the
steady-state ripple content of motor current so as to
minimize certain resulting torque harmonics in the
traction motor. As a general rule, the filter is sized
to limit ripple current to approximately 10 per cent of
the average magnitude of current in the d-c link of the
current fed induction motor drive system.
Insofar as I am presently aware, persons skilled
in the art have not previously recognized that in a traction
vehicle having a self-contained current source a-c motor
drive, the current smoothing filter can be omitted
altogether from the d-c link, thereby realizing a
significant reduction in size, weight, and cost of the
propulsion system, if the d-c generator were replaced
by the rectified output of an a-c generator of the kind
heretofore used for traction vehicle applications. To -
better understand this discovery, the characteristics
of an a-c generator that I believe make it uniquely well
suited to be used in combination with a controlled current
inverter will now be briefly reviewed.
An a-c generator, of the type known as a
synchronous generator, is a machine having an armature
winding in which alternating current flows and a field
winding to which d-c excitation is supplied. The
., .
20-TR-1183
armature winding usually is on the stator, and in a 3-
phase machine the windings of the individual phases are
displaced from each other by 120 electrical degrees in
space around the circumference of the stator-rotor air gap.
The field winding is located on the rotor, which can be
of either salient-pole or cylindrical construction, and
the field poles are excited by direct current brought
in through slip rings or by a brushless exciter. The
field produced by the d-c rotor winding revolves with
the rotor. If the rotor is driven by a prime mover to
which it can be mechanically coupled, the magnetic field
produced by the rotor winding will induce in the stator
windings an alternating voltage having a frequency
proportional to the number of poles and the angular
velocity of the rotor. Ordinarily the stator windings
are distributed in such a manner that the alternating
voltage has a generally sir,usoidal wave form, but they
can alternatively be arranged to generate other wave
forms if desired.
~ynchronous generators designed specifically
for traction vehicle applications (hereinafter referred
to as traction alternators) ordinarily are high-reactance
machines. By high reactance I mean that the armature
reaction (in ampere-turns per pole) of the machine at
full-load current is a large percentage (e.g., 200%) of
the no-load field ampere-turns at rated voltage and
frequency. Armature reaction refers to the effect of
magnetomotive force (mmf) resulting from current in the
_5_ rl
20-TR~1183
~Z ~
armature windings. The armature mmf modifies the
electromagnetic flux produced by current in the field
windings and changes the strength of the resultant field
in the stator-rotor air gap of the machine. If an electric
load connected to the armature terminals of a traction
alternator were short circuited and the field excitation
were held constant, the armature mmf will almost directly
oppose the field mmf, thereby demagnetizing or weakening
the resultant air gap field and limiting the rise in
armature currPnt. In effect the rate of change of current
is limited by a transient reactance similar to the
equivalent reactance of a short-circuited transformer.
The final value of short circuit current is proportional
to the synchronous impedance of the machine.
A typical traction alternator at normal
excitation and speed has sufficiently large reactance to -
limit steady-state short circuit current to less than
rated full-load current. This reactance is sometimes
referred to as the steady-state unsaturated synchronous
reactance XS of the alternator. In practice its per
unit value is in a range from 1.0 to 3.0, and therefore a
typical traction alternator will tend to maintain rated
current with changing load impedance.
Representative of traction alternators suitable
for the practice of my present invention is General
Electric modçl GTA22 manufactured by the General Electric
Company, Transportation Systems Business Division in Erie,
Pennsylvania. The horsepower rating of this machine is
approximately 1,200 at rated engine speed. Contributing
to the relatively high synchronous reactance of the GTA22
20-TR-1183
i1153~Z
alternator is the fact that it has no amortisseur windings.
Amortisseur windings are short-circuited damper bars or
squirrel-cage windings that are often inserted in the
rotor pole faces of synchronous generators for the purpose
of producing torques that help to damp out mechanical
oscillations of the rotor about its equilibrium position.
Such windings also improve the transient voltage regulation
of the machine. By voltage regulation I mean the tendency
of the voltage amplitude at the terminals of the stator
windings to remain substantially constant, under conditions
of constant excitation and frequency, regardless of
variations in the electrical load connected across the
terminals. When amortisseur windings are used in a
synchronous generator, they effectively oppose changes
of electromagnetic flux in the air gap and thereby reduce
the rate at which terminal voltage transiently varies
in response to rapid changes in the load. In some
applications ~e.g., electric utilities where a-c
generators are operated in parallel with one another to
supply interconnected power distribution systems), the
damping and good voltage regulation provided by amortisseur
windings are desired, but in traction applications these
features are not needed, and thus traction alternators
; ordinarily do not require the use of amortisseur windings.
As a result of its high synchronous reactance and its
omission of amortisseur windings, a traction alternator
is specially well suited for use in combination with
controlled current inverters supplying adjustable speed
a-c traction motors.
.
20-TR-1183
11153~Z
Summary of the Invention
~ general objective of the present invention is
to provide an improved traction vehicle propelling process
that can be carried out by lighter weight, more compact,
and less expensive apparatus than has heretofore been
possible.
In carrying out my invention in one form, a
traction vehicle is equipped with a prime mover, at least
one variable-frequency controlled current inverter having
a set of a-c terminals and a pair of d-c terminals, and at
least one adjustable speed a-c traction motor connected to
the a-c terminals of the inverter by way of alternating
current conductors, and the vehicle is propelled (motoring
mode of operation) or retarded (braking mode of operation)
by the process of providing on-board the vehicle a traction
alternator having armature and field windings and a rotor,
drivingly coupling the prime mover to the rotor of the
alternator, exciting the field winding of the alternator,
interconnecting the armature windings of the alternator
and the d-c terminals of the inverter by way of a rectifier
bridge and a d-c link without appreciably smoothing any
direct current flowing in the link, controlling the
switching frequency of the inverter so as to vary, as
desired, the fundamental frequency of the alternating
current energizing the motor during motoring operation of
the vehicle, and controlling the excitation of the field
winding so as to vary, as desired, the fundamental
amplitude of the alternating output current of the
--8--
., .
20-TR-1183
l~S3~Z
alternator armature windings, thereby varying the average
magnitude of direct current in the d-c link and hence the
fundamental amplitude of the alternating current being
supplied to the motor through the aforesaid conductors
S during the motoring mode of operation.
Brief Description of the Drawin~s
My invention will be better understood and its
various objects and advantages will be more fully
appreciated from the following description taken in
conjunction with the accompanying drawings in which:
Fig. 1 is a functional block diagram of a self-
contained propulsion system for a traction vehicle driven
by two sets of inverters and a-c motors interconnected in
series, which system can be used to practice my invention
in one form;
Figs. 2A, 2B, and 2C are charts of input and
output signals of the block labeled Command Logic in Fig.
1, which charts show the relationships between the output
signals and vehicle speed that are typically programmed in
the Command Logic module at two different settings of the
throttle;
Fig. 3 is a graph showing how tractive effort
varies as a function of speed for the Fig. 1 vehicle when
motoring with a throttle setting of 1.0 per unit and also
showing an exemplary tractive effort vs. speed character-
istic during a braking mode of operation;
Fig. 4 is a schematic circuit diagram of a variation
of the power circuit shown in Fig. 1, whereby the traction
motors are energized in parallel instead of in series; and
20-TR-1183
l~S392
Fig. 5 is a schematic diagram of a variation
of one of the Fig. 1 motors and its associated inverter(s)
and controls, which motor utilizes two sets of phase-
displaced star-connected 3-phase stator windir.gs.
Descri~tion of the Preferred Embodiments
In Fig. 1 a traction vehicle such as a locomotive
or an off highway truck is represented generally by the
broken line block 9 having a plurality of wheelsJ only two
of which are shown at 10 and 20. In order to drive the
illustrated wheels and thereby propel or retard the vehicle
9, electric traction motors 11 and 21 are mounted on the
vehicle with their shafts mechanically coupled to the
wheels 10 and 20, respectively. The motors 11 and 21 are
polyphase adjustable speed a-c motors, preferably of the
induction type but optionally of the synchronous, synchronous
reluctance, or other known type, and they can be either
round or linear. Each of the motors 11 and 21 is assumed
to have 3-phase star-connected stator windings that are
connected for energization to a set of a-c terminals of
a corresponding one of a pair of static electric power
inverters 12 and 22 by way of three alternating current
conductors A, B, and C. The number of phases is not
critical, and motors having single, double, six or more
phases can be alternatively used if desired.
Each of the inverters 12 and 22 is a controlled
current inverter suitably constructed and arranged to
excite the stator windings of the associated traction
motor with alternating current of variable frequency
and amplitude. The power circuit shown and described in
--10--
20-TR-1183
11~53~2
U.S. patent No. 3,980,941 - Griebel is well suited for
this purpose. The excitation current is derived from --
a controllable d-c power supply to which the d-c terminals
of each inverter are connected. In accordance with the
present invention, the controllable d-c power supply
comprises a prime mover-driven traction alternator having
armature windings connected to the d-c terminals of the
respective inverters through a rectifier bridge and a d-c
link having no current smoothing choke. In Fig. 1 the
prime mover is shown at 30, the alternator at 34, the
rectifier bridge at 40, and the d-c link at 50.
The prime mover 30 is a thermal machine, such
as a diesel engine or a gas turbine , capable of converting
the heat of a combustible fuel into rotary motion of a
drive shaft. Its rate of rotation (revolutions per
minute) is controlled by a governor 32. In the presently
illustrated embodiment of the system, the prime mover
is intended to run at a variable speed, ~ut alternatively
it could be arranged to run at a substantially constant
speed if desired.
The traction alternator 34 has a rotor that is
mechanically coupled to the drive shaft of the prime mover
30, a field winding 36 located physically on the rotor
and connected electrically to a suitable source 38 of
excitation, and a plurality of armature windings
physically located on the stator of the alternator so
as to have induced therein 3-phase alternating voltage
having a fundamental frequency that varies with the
speed at which the field is rotated and a fundamental
20-TR-1183
1115352
amplitude tha~ depends on both the speed and the
excitation of the field winding 36. The traction
alternator 34 is characterized by a high synchronous
reactance (e.g., approximately 2.0 per unit) and ~he
absence of amortisseur windings. It can advantageously
be the same as or a modification of the General Electric
model GTA22 traction alternator that was referred to in the
introductory portion of this specification. Such a machine
also has a high transient reactance which is advant~geous
in practicing the present invention.
The rectifier bridge 40 comprises an array of
six uncontrolled electric valves or diodes interconnected
and arranged in a full-wave double-way configuration
having three a-c terminals 41, 42, and 43 and a pair of
d-c terminals 45 and 46. The a-c terminals of the
bridge are respectively connected to the armature winding
terminals of the traction alternator 34, and the d-c
terminals are connected to the d-c link 50, whereby the
alternating output current of the alternator 34 is
converted by the bridge 40 to a unidirectional current
in the d-c link 50. The unidirectional current is fed
over the link 50, without being appreciably smoothed or
filtered, to the d-c terminals of the inverters 12 and
22. I have discovered that the heretofore conventional
current smoothing choke can be omitted altogether from
the d-c link 50, or, if provided, its size can be
materially reduced, because of the tendency of the
traction alternator 34 to supply substantially controlled
magnitude current to the inverters 12 and 22, as
~0 previously explained.
-12-
20-TR-1183
l~S392
The d-c link 50 includes a conductor 51
connected between the anode terminal 45 of the bridge
40 and a first d-c terminal 23a of the controlled current
inverter 22, a conductor 52 connected between the other
S d-c terminal 23b of the inverter 22 and a first d-c
terminal 13a of the controlled current inverter 12, and
a conductor 53 connected between the second d-c terminal
13b of the inverter 12 and the cathode terminal 46 of the
bridge 40. There is no voltage smoothing capacitor
spanning the conductors 51 and 53 and no current smoothing
inductor in series therewith. A dynamic braking resistor
54 shunted by a switch 55 is connected in series with one
of these conductors, preferably between the anode terminal
45 of the _ectifier bridge and a proximate terminal 47 of
the conductor 51.
The switch 55 across the resistor 54 is coupled
to a retard controller 56 that determines its open or
closed state. During periods of motoring (M), the switch
55 is closed and thus provides a path of negligible
impedance across the resistor 54 for current flowing in
the d-c link 50 during such periods, whereas during
periods of braking (B) the switch is open and thus
effectively interposes the resistor 54 in series with
the d-c link. When in an electric braking (retarding)
mode of operation, the a-c traction motors 11 and 21 are
driven by the inertia of the vehicle and consequently
serve as generators delivering power to the invertexs 12
and 22, respectively, and during such periods there is a
polarity reversal of the voltage across the inverter d-c
terminals so that the average potential on terminal 23a
"
-13-
.,
- 20-TR-1183
lllS3~Z
(and hence terminal 47 of the d-c link) is negative
rather than positive with respect to terminal 13b.
In this mode the potential diference between the
relatively negative terminal 47 of the d-c link and the
more positive anode terminal 45 of the d-c power supply
will be absorbed by the braking resistor 54, and the
resistor 54 dissipates energy in the form of heat. As
will be understood by persons skilled in the art, the
vehicle controls include suitable means effective when
operating in the retarding mode to reduce the excitation
of the alternator 34 to zero so that none of the diodes
that form the rectifier bridge 40 is reverse biased,
whereby the bridge freely conducts whatever current is
flowing in the d-c link 50 during dynamic braking.
The above-described combination forms a current
fed a-c motor drive system in which each of the controlled
current inverters 12 and 22 is operative, during periods
of motoring, to switch the d-c link current in sequence
between the respective phases of the stator windings of
the connected motor load, thereby supplying to the
associated traction motor fundamental alternating output
Gurrent having a frequency determined by the fundamental
switching frequency of the electric valves in the inverter
and an amplitude determined by the average magnitude of
the unidirectional current in the d-c link 50. The d-c
link current in turn is determined by the amplitude of
the alternating output current of the armature windings
of the traction alternator 34. Because the alternator
has a high reactance and no amortisseur windings, the
. -14-
_ 20-TR-1183
l~llS3~'3Z
amplitude of its output current remains substantially
constant ~assuming the alternator speed and excitation
are constant) although its electrical load impedance
changes widely during the periodic intervals of co~mutation
in the inverters 12 and 22. In other words, the load
impedance imposed by the inverter-motor sets is negligible
compared to the source impedance of the traction alternator,
and the only variable that appreciably influences current
amplitude is the amplitude of alternating voltage induced
~0 in the armature windings of the alternator. Consequently
the rectified output current of the alternator 34 does
; not have to be filtered or smoothed in the d-c link 50.
By appropriately controlling the frequency,
amplitude, and phase sequence of the traction motor
excitation current, the vehicle 9 can be propelled or
retarded in either forward or reverse directions as
desired. For this purpose suitable means is provided for
controlling and regulating the speed and excitation of the
alternator 34 and the switching frequency of the inverters
12 and 22 in programmed response to an operator controlled
throttle 59 (or, during dynamic braking, a manually
operated rheostat 57 associated with the retard controller
56) and to certain feedback signals. The throttle 58
is set in accordance with the motor torque or horsepower ~ -
that is desired during the motoring mode of operation,
while the feedback signals are representative of actual
motor responses. Before proceeding with a detail
description of the illustrated control system, it should
~e emphasized that a vehicle propulsion process embodying
-15-
20-TR-1183
li~lS3~J2
the present invention can alternatively be controlled
or regulated by schemes different than the particular
one that has been shown in Fig. 1 by way of example.
The throttle 58 provides input control signals
to both the governor 32 and a command logic module 60.
The governor 32 is operative in response to changes in
the throttle setting to appropriately adjust the amount
of fuel injection in the prime mover 30 so as to increase
the speed of rotation of the alternator rotor as the
desired horsepower of the vehicle 3 inrreases. As the
rotor speed varies, so does the fundamental amplitude
of both the voltage generated in the armature windings
and the resulting output current of the alternator 34
(assuming that field excitation remains constant and load
impedance is constant or negligible), and in this manner
the magnitude of current supplied to the traction motors
11 and 21 can be varied. A finer or more precise
regulation of motor current is obtained by suitably
controlling the alternator excitation, as is more fully
explained below.
The command logic module 60 is responsive to the
input control signal received from the throttle 58 and to
motor speed feedback signals ~rl and ~ 2 received over
lines 61 and 62 from suitable means, such as tachometer
2S generators 14 and 24, for sensing the actual angular
velocity of the rotors of the motors 11 and 21, respectively.
The module 60 also receives an input control signal over
line 63 from the retard controller 56. Two variable
output signals are derived from this module: on line 64
-16-
20-TR-1183
~1~S392
a first one T~ representing a commanded value of motor
torque; and on line 65 a second one 0C* representing
the desired magnitude of motor excitation. As will soon
be described, the first output signal determines the
fundamental frequency of the traction motor excitation
current, whereas the second output signal determines the
fundamental ~mplitude of this current.
Ordinarily the module 60 will be arranged to
coordinate the value of 0C* with the value of T* and will
be preprogrammed, with conventional rate limits, to vary
the values of both of these output signals during the
motoring mode of operation as functions of the per unit
setting of the throttle 58 in accordance with the schedules
indicated in Figs. 2A, 2B, and 2~ where the abscissa is
scaled in per unit values of vehicle speed, good adhesion
of the wheels being assumed. From the exemplary charts
shown in these figures it will be observed that the first
output signal T* varies directly with the throttle setting
whereas the second output signal 0C* is a square root
function of the throttle setting. In addition, both output
signals vary as the reciprocal of vehicle speed as the
latter increases above a corner point speed of 1.0 per
unit. This results in a constant torque control strategy
below the corn~r point speed and a constant horsepower
control strategy above that speed, analogous respectively
to constant field and to field weakening modes of operation
of a conventional d-c traction motor propulsion system.
If the vehicle 9 were a diesel-electric loco-
motive, the values of the output signals from the command
logic module 60 can be influenced if desired by two other
- 20-TR-1183
S3~2
input signals. One of the optional additional input
signals is received over line 66 from a load control
potentiometer or the like in the governor 32. If, in
attempting to maintain a set point rotational speed of
the prime mover 30, the governor permits fuel to increase
to a specific limit that is predetermined for each throttle
setting, a signal is developed on line 66 to override the
normal program in the command logic module 60 so that the
first outputsignal T* is reduced. This results in
electrically unloading the prime mover to match the fuel
rate that is permitted, and bogging or stalling of the
machine is thereby prevented. The second additional input
signal can be a turbospeed signal applied to a terminal 67,
which signal is representative of the rotor speed of a
turbocharger (not shown) normally associated with the
prime mover 30 to supercharge the air in its intake
manifold. For each throttle setting, if the actual
turbospeed is lower than expected, the turbospeed signal
transiently overrides the normal program in the command
logic module 60 so that the output signal T* tracks the
speed of the turbo while accelerating to steady speed.
As a result the power demand will tend to be compatible
with the amount of air delivered to the combustion chamber
of the prime mover during acceleration, and smoke in the
exhaust is inhibited (see the previously cited U.S.
patent No. 3,878,400).
The first output signal T* on the output line 64
of the command logic module 60 is supplied to summing
means 70 which also receives, on a line 71, a voltage
-18-
lllS:~~Z
~-~ 20 TR 1183
imbalance signal and which produces a compensated torque
command signal representative of the difference between
T* and the voltage imbalance signal (if any). The
compensated torque command signal from the summing means
70 is supplied over a line 72 to a duplicate pair of
inverter control means 15 and 25 that are respectively
associated with the two controlled current inverters
12 and 22.
In the illustrated embodiment of the present
invention, each of the inverter control means 15 and 25
is arranged internally in accordance with the teachings
of United States patent 4,088,934 issued May 9, 1978
in the names of J. D. D'Atre, T. A. Lipo, and A. B.
Plunkett and assigned to the General Electric Company,
and it supplies, on an output line 73, a slip frequency
signal that is used to exert control over the switching
frequency of the associated controlled current inverter
(which frequency determines the fundamental frequency
of the motor excitation current) in a manner to regulate
motor torque and to stabilize the a-c motor drive system.
Toward this end, the first inverter control means 15 is
supplied with a torque feedback signal Tl representative
of the actual magnitude and relative direction of the tor-
que in the rotor of the first motor 11 when excited,
and a torque angle feedback signal sin~Tl representative
of the actual phase angle between current and flux that
interact in the same motor to develop torque. These feed-
back signals are derived from response processor circuits
in a block 74.
-- 19 --
lllS3~2 20 TR 1188
The processor circuits 74 are supplied with a
: first signal il representing excitation current in the
stator windings of the motor 11, as measured by an array
of three current transformers 16 coupled to the alternating
current conductors between this motor and the inverter 12,
and with a second signal ~1 representing the actual electro-
magnetic flux across the stator-rotor air gap in the
mtor 11 as detected by a flux sensor 17 located in this
motor. The response processor circuits 74 also receive
a signal i2 representing excitation current in the stator
of the motor 21, as measured by an array of current
transformers 26 coupled to the conductors between this
motor and the inverter 22, and a signal ~ 2 representing
: the actual flux across the air gap in the motor 21
and detected by a flux sensor 27 located therein, and from
the latter signals they derive and supply to the second
inverter control means 25 a torque feedback signal T2
representative of the actual magnitude and relative direc-
tion of the torque in the~rotor of the second motor 21
when excited and a torque angle feedback sig~al sin~T2
representative of the phase angle between current and
flux that interact in this motor to develop torque.
Preferably the flux sensors 17 and 27 are constructed
in accordance with the teachings of U. S. Patent No.
4,011,489 - issued March 8, 1977 - Franz et al, the torque
processors in block 74 are constructed in accordance with
the teachings of U. S. Patent No. 4,023,083 - May lO, 1977 -
Plunkett and the torque angle processors in the block
74 are constructed in accordance with the teachings of the
above cited United States Patent 4,088,934 - D'Atre et al.
- 20 -
~ 3~2 20 TR 1183
When arranged in the manner just described,
each of the inverter control means 15 and 25 will be
effective to supply on its output line a slip frequency
signal having a substantially constant steady state value
that changes transiently in response to any error between
the actual torque angle of the associated motor, as
represented by the torque angle feedback signal, and a : :~
desired torque angle that itself varies in a correc-
tive sense in response to any non-minimum error between
the associated torque feedback signal (Tl or T2) and
the compensated torque command signal on line 72.
The slip frequency signal on the output line 73 -
of the first inverter control means 15 is combined at a ~
summing point 75 with the first motor speed feedback signal ~- :
~ rl to derive, on a line 76, an excitation frequency control
signal 1~ el* representative of the algebraic sum of the
desired slip frequency and the electrical frequency
equivalent of the actual speed of rotation of the motor 11
shaft. The latter signal is fed to a block 18 symbolizing
known firing logic circuits and a gate pulse generator
for the respective valves of the controlled current
inverter 12, and its value will determine the inverter's
fundamental switching frequency and hence the fundamental
frequency of the polyphase excitation current that is
supplied by the inverter to the first motor 11. The
phase sequence, and hence the direction of the rotation of the
motor 11, corresponds to the sequencing of the gate pulses,
20-TR-1183
11153~2
and it is practically determined by a forward/reverse
command signal applied to the firing logic circuits at
terminal 77. In a similar manner the slip frequency
signal on the output line of the second inverter control
means 25 is combined at a summing point 78 with the second
motor speed feedback signal ~r2 to derive an excitation
frequency control signal ~ 2* representative of their
algebraic sum, which signal is fed over a line 79 ot a block
28 symbolizing known firing logic circuits and a gate
pulse generator for the respective valves of the
controlled ~urrent inverter 22, whereby the inverter's
fundamental switching frequency and hence the fundamental
frequency of the polyphase excitation currents supplied
to the second motor 21 are determined by the value of the
excitation fre~uency control signal on line 79.
In operation, each of the inverter control means
will vary the fundamental excitation frequency of the
associated traction motor as necessary to preserve a
minimum or zero error between desired and actual values
of the torque angle of the motor (i.e., the phase angle
between motor current and motor flux). Consequently the
moments of current transfer or switching in each
controlled current inverter are synchronized with the
counter electromotive force of the motor, and the drive
system is desirably stabilized. By thus stabilizing the
system operation without relying on regulation of current
in the d-c link, there is no need to use fast-responding
power conditioning means such as a chopper or phase controlled
rectifier bridge between the alternator 34 and the d-c
link 50, and there is no need to omit amortisseur windings
in the traction alternator 34.
-22-
20-TR-1183
11~53~32
Each of the inverter control means is also
effective ~o vary the desired torque angle, and hence the
motor excitation frequency, as necessary to minimize or zero
any steady-state error between desired and actual values
of motor torque. The desired value of torque is
determined by the compensated torque command signal on ~ ~
line 72, and it will be the same as the output signal T* ~ -
from the command logic module 60 except when there is a
voltage imbalance signal on line 71. The latter signal
is representative of the difference between the voltage
across the d-c terminals 13a and 13b of the inverter 12
and the voltage across the d-c terminals 23a and 23b of
the inverter 22 when this difference exceeds a predetermined
amount. Before describing the i7 lustrated means for
15 obtaining the voltage imbalance signal, it will be noted ~-
that such means would not be used in any practical
applications ofmy invention that do not require reducing
or reallocating motor torque in the event of une~ual
voltages on the d-c terminals of the respective inverters
12 and 22.
To derive a voltage imbalance signal on the line
71, a pair of conventional voltage transducers 81 and 82
are connected across the d-c terminals of the inverters
12 and 22, respectively, and voltage feedback signals from
these transducers are supplied over the respective lines
83 and 84 to a summing point 85 which produces, on output
line 86, a signal that is a measure of their difference.
The difference signal on line 86 is coupled to the
previously mentioned line 71 through threshold detecting
. ~
20-TR-1183
~l~S392
means 87, and the latter means is effective to prevent any
voltage imbalance signal from appearing on line 71 unless the
difference signal on line 86 is greater than a predetermined
amount in either a positive or a negative sense. For differ-
ences less than the predetermined amount, a deadband existsand no voltage imbalance signal appears on line 71. This
deadband accommodates the normally expected difference between
rolling radii of the respective wheels 10 and 20, and its
si~e will ordinarily be determined by the maximum permissible
speed differential between the two motors 11 and 21. While
not shown in Fig. 1, if desired the voltage imbalance
detecting means could alternatively be arranged selectively
to reduce the torque command for one of the two motors 11 and
21 (the motor whose asEociated inverter has the higher voltage
across its d-c terminals) and to increase correspondingly the
torque command for the other motor (the one whose inverter
has the lower d-c voltage). If the vehicle 9 were an off-
highway truck, means (not shown) can be provided, if desired,
for disabling the threshold detecting means 87 so that no
voltage imbalance signal is present on the line 71 while
the truck is changing its direction of travel or negotiating
turns.
The second output signal 0C* from the command
logic module 60 serves as a reference or command signal
for an alternator excitation control block 90 to which it
is coupled by the line 65. The excitation control block
90 additionally receives an input control signal over line
63 from the brake controller 56 and a motor excitation
magnitude feedback signal that is applied to a terminal
91. The latter signal is representative of the average
-24-
20-TR-1183
1115392
level of excitation in the stators of the traction motors
11 and 21, and during the motoring mode of operation it
advantageously comprises a signal ~av derived by the
response processor circuits 74 as a function of the actual
magnitude of flux across the air gaps in the motors.
The excitation control block 90 is suitably constructed and
arranged to provide, on an output line 92, an alternator
field current signal If* that varies as a function of any
non-minimum difference between the reference signal on
line 65 and the motor excitation magnitude feedback signal
applied to terminal 91. The output signal If* is used to
control the excitation source 38 of the traction alternator
34 so as to determine the magnitude of current in the field
winding 36 which in turn determines the fundamental
: 15 amplitude of the alternating output current of the armature
windings of the alternator 34. The source 38 responds to
variations in the value of If* by varying the amplitude
of the alternator output current as necessary to minimize
or zero the difference between the feedback signal on
terminal 91 and the reference signal 0C*
Any conventional excitation source 38 can be
used in practicing my invention. For example, it could be
a brushless exciter built into the alternator 34, a
separate "static" exciter, or a rotary exciter mechanically
coupled to the drive shaft of the prime mover 30 and
electrically connected through slip rings to the field
winding 36. In the illustrated embodiment of the
invention, field current limiting means is shown between
the output line 92 of the excitation control block 90
and the exciter 38. This means comprises a summing means
-25-
. . .. . . . .... . . . .... . . .. .. . ... .. . . ... . . . . .
20-TR-1183
~1~5392
93 connected to the line 92 for subtracting from the
desired alternator field current signal If* a limit signal
produced by a threshold detector 9~ that is connected to
a current transducer 94 in series with the alternator field
36. The output of the summing means 93 is supplied through
a gain circuit 95 to the exciter 38~ Whenever the actual
magnitude of current in the field winding 36 exceeds a
predetermined maximum limit, the threshold detector 96
appropriately modifies the output signal of the s~mming
means 93 so as to reduce to a safe level the amount of
field current being called for. As a result, alternator
overexcitation is prevented in the event of a malfunction
such as loss of feedback.
In summary, I have disclosed an improved vehicle
propulsion process comprising a new use of a current fed
a-c traction motor drive system wherein the d-c link is
free of current smoothing means. More particularly, the
process includes the steps of providing on-board the
vehicle a traction alternator having armature and field
windings and a rotor, rotating the rotor, exciting the
field winding, converting the alternating output current
of the armature windings to a unidirectional current,
feeding the unidirectional current to d-c terminals of 2
variable-frequency controlleA current inverter without
appreciably smoothing the unidirectional current,
controlling the switching frequency of the inverter so
as to vary the fundamental frequency of its alternating
output current as desired, and controlling the excitation
of the field winding so as to vary, as desired, the
20-TR-1183
`- 111~3~2
fundamental amplitude of the alternator output current
and consequently both the average magnitude of the unidirec-
tional current and the fundamental amplitude of the inverter
output current that energizes an adjustable speed a-c
traction motor.
~ he typical performance of a vehicle using the
above-described process is shown in Fig. 3 where the trace
101 represents the tractive effort vs. speed characteristic
of the vehicle when motoring with a throttle setting of
1.0 per unit. An examplary tractive effort vs. speed
characteristic 102 during a braking or retarding mode of
operation is also shown in Fig. 3. ~lternatively, an
extended braking program, as represented by broken line
103, is possible if the single braking resistor 54 (see
Fig. 1) were replaced by a plurality of sequentially staged
dynamic brake resistors~ Furthermore, if energy were
supplied by the traction alternator, a characteristic
shown by the line 104 can be achieved.
Two variations of the current fed a-c motor
drive system shown in Fig. 1 will now be disclosed.
Instead of being electrically connected in series, the
two a-c traction motors 11 and 21 could alternatively
be paralleled with one another in which case the first
d-c terminals (13a and 23a) of the respective inverters 12
and 22 would be connected in common to the conductor 51
of the d-c link 5~ and the other d-c terminals (13b and 23b)
of the two inverters would be connected in common to the
link conductor 53, as is shown in Fig. 4. In this case it
is desirable to include, in separate connections, relatively
-27-
S3~Z 20-TR-1183
small isolating inductors to prevent the voltage transients
generated during commutation in one inverter from being
coupled into the other inverter and disturbing its normal
operation. This purpose can be served with relatively
small and inexpensive inductors having low values of
inductance (e.g., approximately 10~ of inductance values
of chokes used in the d-c links of prior art current fed
a-c motor drive systems) such that the~ would be
ineffective per se to appreciably smooth the current in the
d-c link. Accordingly, when I use the term "without
appreciably smoothing the unidirectional current" in this
specification and claims I do not intend to exclude the
possibility that decoupling inductors can be added if
desired in the individual lines that connect the d-c link
to the respective inverters in a parallel array of
inverters.
It should be noted that in some practical appli-
cations of the invention, with either the series (Fig. 1)
or parallel (Fig. 4) connections of the inverter-motor
sets, it may be desirable to have the motor excitation
magnitude feedback signal that is applied to terminal 91
of the alternator excitation control block 90 represent
the average magnitude of excitation of whichever motor
has the higher actual magnitude of torque, rather than the
average excitation in both motors.
Turning now to Fig. 5, there is shown a 6-phase
a-c traction motor 111 that can be substituted for each of
the 3-phase motors 11 and 21 shown in Fig. 1. A 6-phase
motor has the advantage of lower torque pulsations
compared to a 3-phase machine. The stator windings of
-28-
20-TR-1183
9Z
the 6-phase motor 111 are actually arranged in two
separate sets lO9a and lO9b of three star-connected
windings, with winding set lO9b being displaced 30
electrical degrees from winding set lO9a. A cGntrolled
S current inverter feeding this motor preferably comprises
two duplicate parts 112a (labeled "INV. ~la" in Fig. 5)
and 112b (labeled "INV. #lb"), each part having the same
power circuit configuration as ~he inverter 12 previously
described. As is indicated in Fig. 5, the d-c terminals
of the two inverters la and lb and an interconnecting
conductor 113 are all serially connected between the
conductors 52 and 53 of the d-c link. The firing logic
and gate pulse circuits 118a and 118b for the respective
inverters la and lb are essentially the same as the
lS corresponding block 18 associated with the inverter 12
of Fig. 1, except that the block 118b is so arranged that
the family of the gate pulses it supplies to the valves of
the inverter lb is time displaced by 30 electrical degrees -
from the gate pulses that the block 118a generates for the
inverter la. Consequently, the inverter lb supplies the
winding set lO9b of the motor 111 with 3-phase alternating
current on lines A', B' and C' having a phase displacement
of 30 degrees with respect to the 3-phase alternating
current that the inverter la supplies on lines A, B and C
to the winding set lO9a. The fundamental frequency of
the alternating current is determined by the value of the
excitation frequency control signals on lines 76a and 76b,
respectively, these signals being derived from the summing
points 75a and 75b as shown in Fig. 5.
-29-
20-TR-1183
, .
While several embodiments of the invention
have been shown and described by way of illustration,
other modifications and variations therein will probably
occur to persons skilled in the art. For example, a
traction alternator having amortisseur windings can be used
in practicing the invention, it being understood that in
this case the peak current in the traction motors during
commutation will be higher than if such windings were
omitted. It is therefore intended by the concluding
claims to cover all such changes and modifications as fall
within the true spirit and scope of this invention.
-30-
