Note: Descriptions are shown in the official language in which they were submitted.
~ 391 21 DSH 2473
The present invention relates generally to power
conversion systems and more particularly to a system for
controlling the tor~ue of an a.c. ~alternating current) motor,
by controlling the torque of the motor through the control
of the mc,tor current, to maximize the motor efficiency
through control of the motor flux, and to provide for dynamic
braking of the motor.
Historically, d.c. (direct current) motors have
been used where operation over a wide speed range is desired.
More recently, a.c. motors have been finding greater
application in var;~a~le speed drive applications. This is
due in large part because of the inherent ruggedness of an
a.c. motor plus a lesser maintenance problem due to the lack
of brushes which make a.c. motors desirable for certain
applications.
There are, howeYer, certain problems associated
with the use of a.c. motors particularly when the motor is
supplied with power from a variable frequency inverter such
as a phase controlled thyristor inverter. This is primarily
the result of the fact that the commutation delay of the
motor current may be as great as 120 degrees. In any phase
controlled conyerter situation, a certain amount of time is
expected between the time of rendering conductive (firing)
one thyristor in the converter until the current will transfer
fxom a preyiously conducting to the newly fired thyristor.
Normally, this overlap of time is from 10 to 30 electrical
degrees. In the case of an inverter situation supplying an
a.c. motor such as an induction motor at high motor speed, it
may take as long as 120 electrical degrees be-tween the firing
of a new thyristor and the transfer of the current to the
motor phase to which the thyristor is connected. As such, the
use of open loop systems in which the thyristors are fired as
-- 1 --
~ 21 ~SH 2~73
a result of a preestablished schedule is not particularly
applicable to this type of operation and instead a closed
loop system in ~hich the thyristors of the inverter are fired
as a function of the exist;ng angle between air gap flux and
stator current (hereinafter referred to as "air gap power
factor"~ represents a more desirable mode of operation.
In addition, because the a.c. current supplied to
the a.c. motor from a current source inverter is not
sinusoidal, the motox tends to exhibit torque pulsations
which can become particularly objectionable at low speeds.
At higher speeds, these pulsations are absorbed by rotor
inertia but at lower speeds, as the inverter approaches zero
frequency, the frequency~ of the torque pulsations can become
lo~ enough to excite mechanical resonances in the structure
or to provide ~hat is generally referred to as cogging in
the motor performance.
It is, therefore, an object o~ the present
in~ention to proyide an improved power conversion system
particularly as applied to the operation of an a.c. motor.
Another object i5 ko provide an improved a.c.
motor control and drive system.
~ furthex object is to provide an improved a.c.
motor control and drive s~tem which operates at high
ef~iciency~
It is an additional object to pxovide improved
a . c. motor control and drive system in which torque ripple,
paxticulaxly at lowex operating speeds, is minimized.
It is a further object to provide an improved
motor control and drive system which can operate at high speed
3a where commutation (overlap) time approaches 120 electrical
degrees.
It is a still further object to provide an
~ 21 DSH 2473
~S~l
improved motor control and drive system including dynamic
braking.
The foregoing and other objects are satisfied in
accordance with the present invention through the provision
of a controlled current inverter system for supplying an a.c.
load, especially a motor, with an a.c. current of variable
magnitude and frequency. The system ~mploys a variable d.c.
power source which is connected to a variable frequency
inverter preferably by way of a d.c. link. Means are included
to develop signals representing the instantaneous electrical
torque of the a.c. motor and the instantaneous angle between
the gap flux and the motor current. Through the establishment
of a torque reference signal, the electric torque signal and
the angle signal are utilized to provide appropriate error
signals, the first serving to control the d.c. current in the
link and the second to control the firing angle of the
inverter with respect to the motor flux such that the air gap
power factor at the motor is controlled. In a more preferred
embodiment, a signal proportional to the gap flux of the
motor is also generated and the difference between this signal
and a reference proportional to the desired flux is utilized
to provide modifying trim signals to each of the two main
paths.
Dynamic braking of the motor is accomplished by
providing a braking resistor which is normally short
circuited but which is put into circuit when such braking is
desired. When operating in the braking mode, the d.c. power
source is effectively short circuited while the torque
reference signal is variously adjusted to provide systemmatic
braking of the motor.
While the present inven-tion is particularly defined
in the claims annexed to and forming a part of this
specification a better understanding can be had from the
~ 3 -
~ 21 DSH 2473
following description taken in conjunction with theaccompanying drawings in which:
Fig. 1 is a schematic block diagram illustrating
the present invention in its preferred embodiment;
Fig. 2, 3 and 4a through 4c are graph wave forms
helpful in understanding the operation of the present invention.
Fig. 5 is a schematic functional diagram
illustrating the dynamic braking feature of the present
invention.
Referring now to Fig. 1 which shows the present
in~ention in its preferred embodiment, it is seen that the more
fundamental aspect of the present invention centers around
the controlled current inverter system. The system includes
a source of variable d.c. current 12 which, in Fig. 1, is
shown as a converter unit ll under the control of a suitable
control means 13. A current (IDC) is supplied from the
source 12 by way of a d.c. link circuit, including a suitable
filter for smoothing the d.c. current from the source 12 such
as an inductor 17, to a suitable inverter circuit 14
2Q including a converter unit 15 under the control of a control
means 16. The output of the inverter circuit 14 is supplied
to a load shown in Fig. l as a motor 18. The d.c. link
circuit also includes a dynamic braking resistor 90 which is
connected in parallel with a normally closed shorting contact
92. The dynamic braking mode of operation of the instant
invention will be discussed later in this specification,
particularly with respect to Fig. 5, and for the present
contact 92 may be considered in its normal closed position
such that resistor 90 is effectively absent from the circuit.
The d.c. source 12 could be any one of a variety
of forms such as, for example, a d.c. chopper whose input
terminals are coupled to an uncontrolled d.c. source. In
~ 21 DSH 2473
such a case the control could be of the time ratio type.
The source 12 could also include some other means for
varying the current emanating from a d.c. source. More
commonly, however, the source 12 would be of the form
illustrated in Fig. 1 in which the conversion unit is a
phase controlled multi-legged bridge, for example a 6-
thyristor bridge, which has its input connected to a three
phase source as represented by terminals Ll, L2 and L3. In
this situation, the control could be of that known type
which is synchronized with the line voltage and which, under
the control of an input signal, varies the output by varying
the firing angles of the bridge rectifiers in response to an
input signal to thereby vary the source output.
A feedback path from the inverter input voltage
(VI) is provided to the d.c. source by way of a filter 19
and a summing junction 20. This is a positive feedback of
unity gain. Ignoring for the moment any other inputs to
summing junction 20, this feedback will cause the output
voltage of unit 11 to match that (VI) reflected by the
inverter 14. With no d.c. voltage across inductor 17, the
inductor will tend to maintain a constant current at any
voltage level set by VI. If a second signal is injected into
summing junction 20; i.e. at terminal 21, this will tend to
cause a voltage across inductor 17 which is proportional to
this second signal. This will cause a rate-of-change of
current (IDC) in inductor 17 which is proportional to the
signal at terminal 21. Thus, the positive feedback signal
through filter 19 has tended to convert d.c. voltage source
12 into a d.c. current source responsive to input at terminal
21; that is, as shown in Fig. 1, the output of an amplifier
52 will be later described.
The inverter circuit 14 includes a suitable
~ 5~ 21 DSH 2473
conversion unit 15 which may also take any suitable form but
which most commonly today would also be a 6-thyristor bridge
as known in the art. The operating frequency of the unit 15
is shown under the control of a control means 16. One well-
known implementation for the control means would include a
voltage controlled oscillator feeding a ring counter the output
signals of which are used to initiate the firing of the
thyristors of the bridgeq In this well-known type of control,
the magnitude to the input signal to the voltage controlled
oscillator controls the inverter output frequency. As is
understood, the instantaneous air gap power factor at the
motor load can be varied by changing the inverter output
frequency since any difference between the frequency of motor
back EMF (flux) and the frequency of inverter current will
appear as a rate of change of phase angle of current with
respect to flux.
Motor 18 is, as was earlier indicated, an a.c.
motor, preferably an a.c. induction motor. As such, as is
well known in the art, the motor will have a stator and a
rotor which is separated from the stator by a gap across which
flux is developed.
In the overall control of the present invention,
four signals find primary application. These signals are,
respectively, proportional to the gap flux ( ~ ), the
electrical torque (T), the angle between the motor flux and
the motor current (e) and the actual motor speed ~N). The
three signals ~ , T and e are derived by suitable calculations
in response to motor operating parameters and are shown
emanating from block 22 in Fig. 1. The exact circuitry of the
block 22 is not of critical importance to the present
invention but it may, for example, be that which is shown
and described in allowed U.S. Patent No. 4.088.934 - dated
-- 6 --
S~-2473
May 9, 1978, by J.T. DIAtra et al, entitled, "Means For
Stabilizing A.C. Electric Motor Drive System" which patent is
assigned to the assignee of the present invention. In
accordance with the description in the aforementioned allowed
U.S. Patent, the block 22 develops the ~ , T and 8 signals as
a function of inputs representing motor parameters including
the motor current (I~) which is shown in Fig. 1 as being
derived from three current sensors 24 associated with the
lines connecting the inverter and the motor 18. The other
inputs to block 22 are signals proportional to motor flux
which are shown as being derived from a pair of flux coils 26
associated with the motor. As illustrated in Fig. 1, in
addition to thelxl, T and ~ signals, block 22 outputs a signal
( /Im /) proportional to the absolute motor current. This
signal represents nothing more than the absolute value of the
rectified and combined individual values of the signals
derived from the three current sensors 24.
A tachometer 30 is shown associated with the motor
18 as indicated by the dashed line 31 and serves as an
exemplary means of providing the fourth or motor speed signal
(N). Tachometer 30 could be any of those well known in the
art, for example, a d.c~ tachometer which will provide a
steady-state output voltage signal proportional to the actual
speed of the motor.
The basic control function of the present
invention has its origin with the establishment of a torque
reference signal. To establish this signal in accordance with
the illustration in Fig. 1, there is shown a suitable means
such as an operator settable rheostat 32 which will output a
signal to a point 33 which is proportional to a desired motor
speed; i.e., a speed reference signal. The speed signal N
~ ~a,~ 21 DSH 2473
from the tachometer 3G is combined with the speed reference
signal from the device 32 in a summing junction 34 such that
the output of junction 34 will be a signal proportional to
the difference between the desired and the actual motor speed.
This difference signal is applied to a suitable amplifier 36
having a transfer function appropriate for speed regulation.
Customarily this transfer function is of an integration type
such as that, for example, expressed by the equation:
Gl = K(l~st) wherein K is a constant,
t is a time constant and S is the ~aplace transform operator.
The output of amplifier 36 at point 38 is a signal which is
designated the torque reference signal. That this actually
is a signal proportional to a desired change in torque is
readily apparent when one considers that, if the speed of
the motor differs from that which is desired, to get to the
desired speed a change in torque will be required.
It will be remembered from the previous discussion
that the purpose of the direct current source 12 was to
provide a d.c. current of varying magnitude in accordance
with the desired torque. The top channel will, therefore,
be discussed first. Since the present invention contemplates
motor operation in both the forward and reverse directions,
and since in either case the torque reference signal at point
38 could be of either relative polarity, this signal is
first applied to an absolute magnitude circuit 40, the output
of which forms one input to a summing junction 42. A second
input to junction 42 is the output of a second absolute
magnitude circuit 44 which has as its input the signal (T)
proportional to the instantaneous torque such that, ignoring
for the present the third input to the summing junction 42,
the output of junction 42 as applied to a suitable amplifier
46 is a signal proportional to the difference between the
-- 8 --
~5~ 21 DS~ 247~
torque reference signal and the actual instantaneous torqu~.
The output of amplifier 46 is applied to a limiting circuit
or limiter 48 which serves to prevent exceptionally large
excursions from occurring in short periods of time and thus
limits the rates of which change can be effected. The output
of limiter 48 is applied by way of a summing junction 50 which
has as a second input a signal proportional to the absolute
magnitude of the motor current ( ¦IM¦ ) from block 22. This
particular feedback of the ¦IM/ signal, shown only for
purposes of completeness in the overall system, serves as a
small negative feedback in a stabilizing airection so that when
a large torque is called for, the torque error may be satisfied
transiently by a large current rather than a large torque
until the actual torque can build up. As such, the output
from the summing junction 50 is supplied to a suitable
amplifier 52 the output of which serves as an input to terminal
21 of the summing junction 20 previously discussed. Since,
as pre~iously indicated, the feedback (VI) from the
inverter voltag~ by way of the filter 19 causes the current
IDC to be controlled by the signal at terminal 21, errors in
torque will be corrected through the upper channel path to
the control 13 such that when more torque is required a
positive voltage will be passed to the control 13 to effect
more current fro~ the source 12. Conversely, when less
torque is required, a negative or less positive signal will be
applied to the control 13 thus decreasing the amount of
current supplied from the source to the inverter 14.
The next path to be discussed is the lower control
path on the diagram of Fig. 1. This is the path which controls
the frequency of inverter operation and hence the air gap
po~er factor. Looking at this lower path, it is first seen
that the speed signal N, from tachometer 30, forms a positive
_ g _
21 ~SH 2473
feedback which is analogous to the feedback for the d.c~ link
voltage to the source 12 by way of the filter 19. That is,
the tachometer feeds back a d.c. signal in a positive sense
which commands the inverter frequency to remain at the zero
slip value. As such, the rest of the lower channel has only
to handle signals proportional to slip frequency.
It was earlier indicated that the inverter 14 is
controlled to provide a specified air gap power factor. This
power factor may be defined by an angle ~ and before continuing
with the discussion of the lower control channel, therefore,
it is believed desirable to define what is meant by the angle
~. Referring to the graph of Fig. 2, the abscissa is labeled
the flux axis and the ordinate the voltage axis. The induced
motor voltage is shown as EM. Also shown is the motor
current IM which may be considered as two components, a direct
portion Id which is in phase with the flux and a quadrature
current Iq which is 90 degrees leading the current Id. As is
well known in the motor art, the current Iq is the one which
produces torque or power while the reactive component of the
motor current, Id, is that which produces flux. As such, for
any particular, motor there is a definable relationship
between Iq and Id which will provide the optimum operation of
the motor from a current utilization standpoint. This
relationship may also be defined by the angle ~ between IM
and Id and if, for a particular motor, this angle is held
constant, the motor will run at a constant power factor any
time that it is loaded. Thus, the power factor can be
optimized for the motor.
With the foregoing in mind, beginning again at
point 38 the torque reference signal is applied to a suitable
limiting circuit or limiter 60 which provides, essentially,
an output signal of constant magnitude but of varying polarity
~,~, ]. O
,,, j,,
~ 21 ~SH 2473
in accordance with the polarity of the torque reference signal.
This output of the limiter 60, an angle reference signal
proportional to the desired value of e, is applied to a
multiplier 62 which for the moment may be considered as having
a second input of unity such that its output is an exact
duplicate of its input. The output of the multiplier 62 is
applied as one input to a summing junction 64 the other input
to which is the angle signal 0 ~rom the block 22. These two
signals, the angle reference signal from block 60 and the angle
signal from block 22 form a frequency error signal which is
applied by way of a suitable amplifier 66 to the summing
junction 54. As previously discussed, the output of junction
54 is a frequency command signal which is applied to the
control 16 to thereby control the frequency of the output of
the inverter 14. (It is noted, referencing block 60, that if
the torque reference is zero the angle reference signal from
the angle program 60 is also zero. This is the no load
condition at which the power factor can only be zero and,
therefore, angle e must be zero as well). From the preceding,
the lower loop is essentially a phase-locked loop which senses
an angle error to control the frequency and hence the air gap
power factor of the motor.
From the description thus far it is seen that by
holding the air gap power factor of the motor constant
(bottom channel) and controlling the instantaneous eurrent
through the top channel, a preeise and instantaneous control
of both torque and flux in the motor is maintained. It is
noted, however, that this preeise control ean only be obtained
if the motor eharacteristics are very constant, linear, and
are accurately known such that the angle e can be calculated
very precisely. Since this is not ordinarily the case, the
present invention provides a third eontrol loop. This third
-- 11 --
~ 39~ 21 DSH 2473
loop employs the, ~ signal from block 22 and functions as a
low gain trim to provide modifications to both the currént
control and frequency control paths to assure that the motor
flux is at the proper value for each torque level.
Prior to the discussion of this third path,
however, an understanding of the representation of Fig. 3 is
believed advantageous. In Fig. 3 there is shown the well-known
characteristic in which current or flux is plotted as a ',
function of torque. As is seen from the solid-line depiction ',
of Fig. 3, when the motor is operated at constant angle
(or constant slip) the current of flux is proportional to
the square root of the torque. The characteristic thus takes
on the typical parabolic shape shown. As will become evident
as the description proceeds~ however, it is not desirable to
ever allow the flux tc go to zero in order to keep the angle
loop active at no load and to be ready to deliver torque
quickly if called upon. To this end,,a desirable
characteristic is that which is shown in Fig. 3 by the
dashed lines which modify the basic parabolic function for
both forward and reverse directions.
Returning again to Fig. 1, block 70 represents a
flux program which receives as its input the torque reference
signal at point 38. The similarity between the depicted
function of block 70 and the dashed line graph of Fig. 3 is
noted and it is seen that this may be achieved by a simple
magnitude limited absolute ~alue circuit having an offset
such that the output of the block 70 always has some finite
value even though the signal at point 38 goes to zero. The
output from block 70, a flux reference signal, is supplied
as one input to a summing junction 72 the other input of which
is the ~ signal (from block 22) which has a magnitude
proportional to the instantaneous value of the motor gap flux.
~ 9~ 21 DSH 2473
The output from the summing junction 72 is here termed a
flux error signal and appears at junction 75. The output
from summing junction 72 is applied by way of a suitable
amplifier 74 to the summing junction 42 in a positive sense.
This positive addition to the torque control channel has the
effect that if the instantaneous flux is less than that
desired, the addition at summing junction 42 will result
in an increase in current to the inverter 14. Conversely
if the flux is higher than desired, a smaller current will be
supplied to the inverter.
The output of the summing junction 72 is also
prov~ded to a simple limiter 76 which is provided with an
offset so that with a zero value of the flux error signal
at junction 75 the output of the function 76, after being
applied through a suitable gain amplifier 78 the output of
which serves as an input to the multiplier 62 previously
mentioned, is unity. Other values of the flux error signal
will result in a multiplication factor of greater or less
than unity but always greater than zero being applied to the
multiplier 62. The effect of this flux error signal on the
angle command is to reduce angle when more flux is needed.
The overall effect of the middle or flux loop as
a modifier to the other two loops is essentially as follows.
When there is no flux error, there is no effect from the flux
loop on the multiplier 62 and, hence, the output of this
multiplier is a function of its quiescent gain. When there
is a flux error, the path including the limiter 76 and
amplifier 78 serves to change the gain of the multiplier 62
to correct that error. At very light motor loads the output
of summing junction 50 in the torque regulation loop (upper
loop) might be insufficient to maintain predictable control
of the flux and the feedback path for the 1ux error by way
- 13 -
- .
21 DSH 247~
of amplifier 74 serves to control this flux level at these
light loads by the current control previously mentioned.
Thus, when the motor is unloaded, the motor slip and angle
e are both near zero. The flux control path through limiter
76 and amplifier 78 which acts on motor frequency (angle e~
cannot control flux. The path of the flux error signal by
way of amplifier 76 can act to control flux by controlling
motor current at no load. In the heavily loaded condition
by making the gain of amplifier 74 low, its output will be
small compared to the torque signals through blocks 40 and
44 so that the flux loop will have little effect on current.
Thus, when the motor is loaded the flux error is kept small
by the action of path through multiplier 62 and the path by
way of amplifier 74 has little effect. When the motor load
is light and the flux cannot be controlled through the
frequency, the flux error becomes large enough for response
through the summing junction 42 in the torque regulation
path.
It was earlier mentioned that torque pulsations
tend to be a problem in the type of motor drive here being
discussed and that these pulsations can become particularly
bothersome at low operating speeds. One way to understand
the cause of the torque pulsations is to consider the power
flow through the in~erter and motor. If it is assumed, as
may be reasonably done, that the power coming from the shaft
of the motor is instantaneously equal to the power -that is
applied to the inverter, then it is seen that the shaft power
at the motsr output is approximately equal to the produce of
the d.c. current (IDC~ and the inverter reflected voltage VI.
This conclusion is based upon the assumption that the shaft
speed does not significantly change such that the shaft torque
repXesents the sha~t power. It also assumes that there ~s no
~ 21 DSH ~473
stored energy in either the inverter or the motor.
Based upon these assumptions it is apparent that, because
of the nature of the inverter, there will be a voltage ripple
which occurs at the input of the inverter which will be seen
as a torque ripple at the shaft output. This ripple is
illustrated in Fig. 4a. If Fig. 4a represents the voltage at
the inverter input and the d.c. current IDC is steady-state,
then it is apparent that there will be torque ripple at the
motor shaft output which is also represented by Fig. 4a. It
is an established fact that little can be done to control the
inverter voltage to remove the ripple. The same, however,
does not necessarily hold true for the current. As such, if
the current were to have the appearance shown in Fig. 4b by
the solid line; that is, a waveshape which is essentially the
inverse of the depiction of Fig. 4a, then it is readily apparent
that the power input to the inverter and hence the torque
output of the motor would be a steady-state. As a practical
matter, the idealized current form shown by this solid line of
Fig. 4b is not easily achieved short of a very elaborate
anticipatory program. The control of Fig. l is a reaction
type control and will notl therefore, produce the solid line
waveshape of Fig. 4b. It will, however, because of the
parameters sensed and in accordance with the earlier
description produce a current waveshape similar to that shown
by the dashed line in Fig. 4b. If then the dotted line
representing current of Fig. 4b is multiplied by the
depiction of Fig. 4a, the result representing the
instantaneous torque output of the motor will be something
which can be represented basically by that shown in Fig. 4c.
That is, the torque would be substantially constant but with
small, sharp peaks. These peaks are in no way desirable and
are less than the ideal but the Fig. 4c depiction does
~ 21 DS~ 2473
represent a considerable improvement over the torque
representation as shown in Fig. 4a. The present invention,
as represented by the preferred embodiment showing of Fig. 1
achieves this result.
Thus, it is seen from the description thus far
that there has been provided a power conversion scheme
particularly adapted for motor ~,ontrol which is economically
feasible and which provides precise control of instantaneous
torque for fast dynamic response and for reduction of torque
pulsations particularly at low speed.
Previous mention was made of the dynamic braking
mode of operation which also forms a feature of the present
invention. It will be remembered that, in the early part of
the description regarding Fig. 1, a resistor 90 having a short-
circuiting contactor 92 was provided in the d.c. link circuit
connecting the source 12 with the inverter 14. Fig. 5 taken
in conjunction with Fig. l illustrates how the resistor is
employed, in conjunction with the control of Fig. 1 to effect
the breaking mode of operation in the overall invention.
Referencing now Fig. 5, the three phase source,
again represented by terminals Ll, L2 and L3 is shown
connected to a signal level rectifier 94 such that the output
of the rectifier is a signal having level proportional to the
source voltage. This signal is applied as one input to a
simple voltage comparator 96, for example an operational
amplifier connected in the voltage comparison mode. The
second input to the comparator 96 is shown as being derived
from the wiper arm of a potentiometer 98 which is connected
between a source of positive potential (+V) and ground.
3~ Potentiometer 98 represents any sui-table means for developing
a signal representing a safe operational level of the motor
dxi~e of Fig. l. This level may be, for example, at sevent~
- 16 -
3~3 Zl ~SH ~!47~
percent of the normal voltage at terminals Ll, L2 and L3~
As such, in the embodiment illustrated, comparator 96 will
provide a high level signal only when the voltage from
potentiometer 96 exceeds that from the rectifier 94~
The output from comparator 96 forms one input to
an OR function circuit represented by gate 100 which has a
second input connected to an additional source of positive
potential (+V') by way of a switch 102. This portion of the
depiction is included solely for purposes of illustrating a
complete typical system and as such switch 102 represents a
command source for an "emergency stop". Thus, when either
comparator 96 is outputting a high level signal or when
switch 102 is closed, gate 100 will present a high level
signal on a bus 104 which signal, herein designated a "braking
signal", specifies that the system of Fig. 1 will enter into
the dynamic braking mode of operation.
As shown in Fig. 5, the braking signal on bus 104
is simultaneously applied to several channels to effect various
operations. Firstly, this braking signal acts to force the
speed reference signal (point 33 of Figs. 1 and 5) to a
value specifying zero speed. In Fig. 5 this is illustrated
by the connection of point 33 to ground by way of a suitable
switching means shown as a field effect transistor (FET) 10~.
if FET 106 is the enhancement mode, the application of the
positive braking signal to its base will place the FET into
conduction and thus pull point 33 to ground potential. A
second application of the braking signal on bus 104 is to the
control 13 of the d.c. source 12. Control 13 is so constructed
and arranged that the receipt of a positive signal by way of
bus 104 will result in the short circuiting of the output of
unit 11. The manner in which this short circuiting occurs
will, of course, depend upon the nature of unit ll and the
- 17 -
~ H ~47~
means provided but if, as an example, the unit were a phase
controlled multi-legged thyristor bridge as suggested in the
preferred embodiment, short circuiting could be achieved by
simultaneously firing all thyristors in a leg.
The third application of the braking signal on
bus 104 is to the base of an additional switch, shown as
a FET 108, by way of a normally closed contactor 110. The
source-drain circuit of FET 108 connects point 38 (Fig.l~
to a value representing a demand for zero torque (e.g.,
ground) in the same manner as FET 106 connects point 33 to
ground.
The last employment of the signal on blus 104 is
to operate the contacts 92 and 110. This is symbolically
illustrated in Fig. 5 by the application of this signal to
a coil 112 such that the coil is energized to effect the
opening of the two contacts 110 and 92, thus removing the
forced zero torque reference signal and inserting the resistor
90 into the d.c. link circuit (Fig.l).
The operation of ~he Fig. 5 showing is believed
ZQ apparent from the preceding but may be briefly summarized as
follows. ~7hen dynamic braking is desired, the braking signal
is placed on bus 104 which immediately effects the forcing of
the speed and torque reference signals to levels
representing zero demand and the d.c. source is short circuited.
The reaction time of coil 112 presents a slight delay and
after that delay the braking resistor is placed in circuit
while, simultaneously, the torque reference signal, by virtue
of the opening of contact 110 is allowed to assume a new value
as a function of the forced zero speed references signal and
the extant values of the speed signal N (that is, as a
function of the output of summing junction 34 of Fig. 1) and
the motor is brought to a controlled stop under the full
/
- 18 ~
~ DSH-~47
control of the flux ( ~) and angle (~) loops.
Thus, it is seen that the system of the present
invention provides not only for the control earlier described
but also for orderly entry into a dynamic braking mode of
operation.
While there have been shown and described what is
at present considered to be the preferred embodiment of the
present invention, modifications thereto will readily occur to
those skilled in the art. It is not desired, therefore, that
the invention be limited to the specific arrangement shown and
described and it is intended to cover in the appended claims
all such modifications as fall within the true spirit and
scope of the invention.
., .,~ . -- 1 9
,~
