U.S. patent number 3,743,921 [Application Number 05/262,122] was granted by the patent office on 1973-07-03 for tap changing current regulator.
Invention is credited to Eric G. Cowie, Brian C. Legg.
United States Patent |
3,743,921 |
Legg , et al. |
July 3, 1973 |
TAP CHANGING CURRENT REGULATOR
Abstract
A constant current control system for lighting applications, in
particular airport series lighting arrangements, using a multi-tap
transformer connected to the load effectively through back-to-back
thyristors which are selectively fired to connect the various taps
to the load in order to regulate the current. A signal
representative of the square of the current is used to produce the
control signal which controls the firing point of the various
thyristors. The control signal is a series of impulses produced by
a multivibrator.
Inventors: |
Legg; Brian C. (Hamilton,
Ontario, CA), Cowie; Eric G. (Burlington, Ontario,
CA) |
Family
ID: |
4090203 |
Appl.
No.: |
05/262,122 |
Filed: |
June 12, 1972 |
Foreign Application Priority Data
Current U.S.
Class: |
323/258; 315/279;
315/219; 315/307 |
Current CPC
Class: |
G05F
1/20 (20130101); H05B 39/044 (20130101); H05B
39/041 (20130101); G05F 1/445 (20130101); Y02B
20/146 (20130101); Y02B 20/00 (20130101) |
Current International
Class: |
H05B
39/04 (20060101); H05B 39/00 (20060101); G05F
1/10 (20060101); G05F 1/20 (20060101); G05F
1/445 (20060101); G05f 001/20 () |
Field of
Search: |
;315/196,219,251,258,279,307 ;323/4,6,7,24,43.5S,45,62 |
References Cited
[Referenced By]
U.S. Patent Documents
Primary Examiner: Pellinen; A. D.
Claims
The embodiments of the invention in which an exclusive property or
privilege is claimed are defined as follows:
1. An alternating current regulation system comprising a supply
transformer supplied with alternating current having at least two
secondary taps, means to selectively connect a load to one of said
taps during portions of the cycle and thereby regulate the current
supplied to said load, said means including a pair of
oppositely-poled thyristors connected to each tap, means to produce
a first signal representative of the square of the currents
supplied to the load, means to produce a second signal
representative of the desired load current, means to compare said
first and second signals to produce an error signal, means to
integrate said error signal, means to compare said integrated error
signal with a first reference signal and produce a first output
signal when said error signal exceeds said first reference signal,
a source of cyclic sawtooth waveforms in phase synchronism with the
alternating current being regulated, means to combine said output
signal with said sawtooth waveform and produce a first control
signal when said sawtooth waveform is equal in absolute value to
said output signal, a first oscillator rendered operative by said
first control signal, means to apply the output of said first
oscillator to the firing electrodes of a first of said pairs of
oppositely-poled thyristors in the proper phase to cause the proper
one of said first pair of thyristors to conduct and connect a first
tap to said load, means to compare said integrated error signal
with a second reference signal and produce a second output signal
when said error signal exceeds said second reference signal, means
to produce a second control signal, a second oscillator rendered
operative by said second control signal and means to apply the
output of said second oscillator to the firing electrodes of a
second of said pairs of oppositely-poled thyristors in the proper
phase to cause the proper one of said second pair of thyristors to
conduct and connect a second tap to said load.
2. A regulation system as claimed in claim 1 wherein said second
reference signal is greater than said first reference signal and
the voltage supplied by said second tap is greater than the voltage
applied by said first tap.
Description
BACKGROUND OF THE INVENTION
This invention relates to the regulation of power supplied to a
load, in particular a series lighting load such as a lighting
arrangement conventionally used for airport lighting.
In the prior control of airport lighting, it has been conventional
to use saturable reactors to control the current supplied to the
series-arranged lamps. It is conventional in the operation of such
lamps that a short circuit be placed across the lamps when they are
not operative. When the supply is applied to this load in its short
circuit condition, it will be evident that, in the absence of
regulation, prohibitively large currents would flow. However, in
the prior art it was usual, as has been previously indicated, to
unsaturate the reactor under these conditions and cause the short
circuit to appear, so far as the actual source of energy is
concerned, as a highly reactive load. The current was then limited
in accordance with the reactance to the permissible current limit.
The resultant power would be essentially wholly reactive and would
represent the source voltage times the reactive current in VARS.
Such a reactive load is obviously disadvantageous particularly in
view of power factor penalties which are assessable by suppliers
and the excessively high line currents when supplied by standby
Diesel generator sets. It would therefore be desirable to reduce
this if possible. Saturable reactors also had disadvantages in
their slow response time, high losses and severe waveform
distortion would could give rise to undesirable harmonics and
radiation, which are particularly objectionable in the vicinity of
an airport.
SUMMARY OF THE INVENTION
These and other objections are overcome in the present invention by
utilizing solid state switching devices, not only to regulate the
current by controlling the conducting angle of the solid state
devices, but also selectively to supply current from different taps
of a supply transformer in accordance with the current demand so
that the wave shape is more closely related to sinusoidal, and also
that in short circuit conditions a lower voltage is applied from
the source, thus reducing the total VAR consumption of the system
under these short circuit conditions.
A clearer understanding of our invention may be had from a
consideration of the following drawings, in which:
FIG. 1 is a schematic diagram of a system in accordance with our
invention; and
FIG. 2 is a waveform diagram useful in explaining the operation of
the circuit.
DESCRIPTION OF THE PREFERRED EMBODIMENTS
As will be seen in FIG. 1, the load transformer 10 is supplied from
a source of 60-cycle alternating current at the available voltage
as convenient. The secondary of this transformer is provided with
several taps. Connected to one of these taps, designated 11, are a
pair of thyristors 12 and 13 arranged back-to-back. The other ends
of these thyristors are connected together and to the primary of
the output transformer 14. A similar pair of thyristors designated
15 and 16 are connected to tap 17 and to the primary of output
transformer 14. To the output of the output transformer 14 is
connected the load, consisting of a series of lamps 18, 18a, 18b,
etc., each supplied through its individual transformer 22, 22A,
22B, etc. This load is connected to the secondary of the
transformer through a series of contacts 20a, 20b, 20c and 20d, all
of which are interrelated so that when 20a and 20b are closed, 20d
and 20c are open and conversely when 20d and 20c are closed, 20a
and 20b are open.
A current transformer 19 is connected in series with the primary of
transformer 14 and the output of this transformer is applied to
transducer 21. Transducer 21 is arranged to produce a positive
potential representative of the square of the current passing
through the current transformer 19. This signal is passed through
resistor 24 to the operational amplifier 28.
This operational amplifier, together with all similar operational
amplifiers in the system, is provided with the usual feedback
circuits and potentials to permit its normal operation, all of
which are not shown. Also applied to the operational amplifier 28
is a reference signal derived from potentiometer 26 which is
negative in polarity and is applied through resistor 27 to the
input of the operational amplifier. This reference signal
establishes the desired brilliance of the lamps 18. As will be
seen, the input to the operational amplifier 28 is the sum of a
positive and a negative signal, and therefore represents the
difference between the output from transducer 21 and the reference,
which may be referred to as an error. This error is integrated in
amplifier 28 which is not a simple operational amplifier, but is an
integrating amplifier. As a result, its output represents the
accumulated error. This output is applied to a further operational
amplifier 29, whose point of operation is established by the bias
provided from potentiometer 30. When and only when the signal from
the operational amplifier 28 exceeds the bias potential, amplifier
29 produces an output which is applied through resistor 33 to
operational amplifier 31.
A further signal is applied to operational amplifier 31 from the
ramp generator 32 which is provided with a reference frequency at
60 cycles, which corresponds to the source. In response to this
reference, the ramp generator produces a linearly decreasing
sawtooth wave, as shown at C in FIG. 2, in phase synchronism with
the waveform of the source and having a duration of one-half cycle.
This ramp is applied through resistor 34 to amplifier 31. Amplifier
31 is arranged so that it provides at its output as long as the
combination of the interated signal and the ramp is negative, a
positive output.
The next portion of the circuit consists of an operational
amplifier arranged as a multivibrator, controlled by the potential
at the junction of resistors 35 and 36. Operational amplifier 37
and its associated components operate as a multivibrator, producing
a series of square waves at the output terminal which is connected
to resistor 38. The mode of operation of the multivibrator will be
explained in greater detail subsequently, but it will be noted that
the output is connected to the input through resistors 39 and 40
and diodes 41 and 42. The input of the amplifier is also connected
to ground through capacitor 43 and across the capacitor is
connected a diode 44 and a Zener diode 45 in series. The other
input of the amplifier, that is, the non-inverting input, is
connected through resistor 48 to a potentiometer consisting of
resistors 46 and 47 which are connected between the output of the
amplifier and ground. This input terminal of the amplifier is also
connected through resistor 49 and diode 77 to the junction of
resistors 35 and 36. The square waves produced by the amplifier 37
functioning as a multivibrator are applied through resistor 38 to
the pulse forming network which is a symmetrical circuit arranged
to operated on alternate half-cycles and thereby provide the
necessary controls to the gates of thyristors 15 and 16.
As will be seen, the resistor 38 is connected through resistors 50
and 51 to the bases of transistors 52 and 53. The collectors of
these transistors are connected to a source of positive potential
at terminal 54 through resistors 55 and 56. The collectors are also
connected to the collectors of switching transistors 57 and 58; the
emitters of both of these transistors being connected to ground.
The emitters of transistors 52 and 53 are connected to the bases of
transistors 59 and 60. The collectors of these transistors are
connected to the positive supply terminal 54 through resistors 61
and 62 and load resistors 63 and 64. The output from the
transistors is applied from the load resistors to the primary of
pulse transformers 65 and 66. The emitters of transistors 59 and 60
are connected to ground. The bases of transistors 57 and 58 are
driven from switching transformer 67. One terminal of this
transformer is connected through diode 68 and resistor 69 to the
base of transistor 57. The base of transistor 57 is also connected
to ground through resistor 70 and Zener diode 71. In a similar
manner, the opposite terminal of transformer 67 is connected
through diode 72 and resistor 73 to the base of transistor 58,
which base is connected to ground through Zener diode 74 and
resistor 75. The centre point of the secondary transformer 67 is
connected to the positive supply terminal 54 and the primary is
provided with 60-cycle alternating current as indicated; this
current being from the same source as other 60-cycle supplies in
the system. The junction point of resistors 61 and 63 and resistors
62 and 64 is by-passed to ground through capacitors 77 and 76
respectively. The secondary of transformer 66 is connected between
the gate and cathode of thyristor 16 and the secondary of
transformer 65 is connected between the gate and cathode of
thyristor 15.
The firing circuit just described is associated with thyristors 15
and 16. A similar circuit is of course required in association with
thyristors 12 and 13, which circuit corresponds exactly with that
shown and utilizes, insofar as possible, common waveforms and
components. The basis of differentiation between the circuits is
the bias supplied to the operational amplifier 28 and the
corresponding amplifiers in further firing circuits.
OPERATION
Let us assume that a 60-cycle supply, as shown at A in FIG. 2, is
provided for transformer 10 and the system is in operation
providing current to the load consisting of lamps 18, 18a, etc.
with the switches in the condition shown producing an effective
load resistance as shown at time period t.sub.1 of curve B in FIG.
2. Waveform C in FIG. 2 is intended to be illustrative only. It
will be seen that the solid line represents the ramp, while the
dotted line represents the integrated output from amplifier 29. The
polarity of this latter should be disregarded however, since it
effectively illustrates the zero potential level of the combined
signal indicating the point of transition from positive to negative
value. At the point of intersection, the potential at the input of
amplifier 31 changes polarity. It will therefore be seen that the
polarity of the output from amplifier 31 varies from negative to
positive at a time determined by the combination of the ramp and
the integrated error signal.
Let us now consider the operation of the multivibrator consisting
of amplifier 37 and its associated components. Let us assume that
the output terminal of amplifier 37 is in its most positive state
and has been in this state long enough to charge up capacitor 43 to
a point where the inverting input terminal of the amplifier
connected to capacitor 43 has just reached the potential of the
non-inverting input terminal connected to resistor 49. The output
will suddenly decrease and the output terminal of the amplifier
will go to a negative value depending on its design. It remains in
this condition until capacitor 43 discharges through diode 41 and
resistor 39. When capacitor 43 has discharged to a sufficiently low
level, the amplifier once again reverses its condition and the
output becomes positive. This reversal of positive to negative
condition occurs repetitively as long as the non-inverting input
terminal of amplifier 37 is permitted to go positive, and occurs at
a periodicity determined by resistors 39 and 40. It will be noted
that one resistor, i.e. 39, determines the discharge rate and
another resistor 40 determines the charge rate because of the
inclusion of diodes 41 and 42.
If, however, the potential at the junction of resistors 35 and 36
is negative, the non-inverting input terminal to amplifier 37 is
held at a negative potential. This negative potential at the
non-inverting input to amplifier 37 produces a negative potential
at the amplifier output, which charges capacitor 43 to a negative
potential through resistor 39 and diode 41. The charge on capacitor
43 is limited by diode 44 and Zener diode 45 so that the magnitude
of the potential at the non-inverting input is greater than that at
the inverting input and the multivibrator action is arrested. In
the absence of Zener diode 45, the capacitor would continue
charging until it reaches the amplifier saturation voltage which
would prevent the multivibrator resuming operation in a normal
manner. As soon as the output from amplifier 31 goes positive, the
multivibrator circuit is permitted to operate and produces a series
of pulses. These pulses continue as long as the output from
amplifier 31 is positive. The output pulses are applied as
previously indicated through resistor 38 to the pulse forming
network. The operation of the pulse forming network is as
follows.
Assuming a series of pulses is applied through resistor 38 to the
base of transistor 52 for example, this would normally cause
transistor 52 to become conductive. However, it will be noted that
transistor 57 is placed between the collector and ground on
transistor 52 so that if transistor 57 is switched on, transistor
52 cannot be switched on. The operation of transistor 57 is
determined by the switching circuit which includes transformer 67.
It will be seen that the centre tap of the secondary of this
transformer is supplied with a positive potential. In addition, let
us assume that the secondary of the transformer is providing
positive potential to diode 68 through resistor 69 to the base of
transistor 57. This will cause transistor 57 to be conductive and
transistor 52 therefore can not be operative. Zener diode 71 is
provided to protect transistor 57 from the voltages produced by the
combination of the supply voltage from terminal 54 and the output
from the secondary of transformer 67. In order to provide rapid
switching it is desirable that the output from transformer 67 be
large thus producing an abrupt rise of voltage to a point where
transistor 57 becomes conductive. On the other hand, it is not
desired that large voltages be applied to the base of the
transistor and these voltages are instead absorbed through the
Zener diode 71.
When the polarity of the secondary of transformer 67 reverses, a
negative potential appears on its secondary. This causes transistor
57 to be cut off and in this condition transistor 52 can become
operative. With the pulses applied to its base through resistor 50,
it then applies a similar series of pulses to the base of
transistor 59. This transistor in turn conducts a series of current
pulses which appears across its load resistor 63 and is applied to
pulse transformer 65. These pulses of the form shown at C' in FIG.
3 in turn are applied to the gate and cathode of thyristor 15
causing the thyristor to commence conducting.
It will be seen therefore that the point of conduction of the
thyristor is determined by the time of conduction of transistor 52,
which can operate only during a particular half-cycle which is
arranged to be the conductive half-cycle of thyristor 15, and also
during the period which the multivibrator circuit produces pulses,
this period being determined in time by the error signal and its
point of coincidence with the ramp produced by the ramp generator.
It will therefore appear that the point of conduction of thyristor
15 is determined by the current passing through current transformer
19. In a similar manner, the conducting period of thyristor 16 is
also so determined. Thyristors 12 and 13 are similarly controlled
by a similar circuit, the only difference being the adjustment of
potentiometer corresponding to potentiometer 30, which determines
the thyristor to be connected depending on the current required.
Under circumstances where the load is not connected as shown, but
instead is in its non-operative condition; that is, with contacts
20a and 20b open, and contacts 20d and 20c closed, a short circuit
is applied to the secondary of transformer 14. This causes a
current in the current transformer 19 whch rapidly rises to and
exceeds the value established by potentiometer 26.
As shown by the dotted line in curve C of FIG. 2, the integrated
error signal changes rapidly in a positive direction so that the
effective point of intersection between the ramp and the integrated
signal is delayed to a point late in the cycle. Therefore, the
effective potential at the input of amplifier 31 only becomes
negative late in the cycle and therefore the point of conduction of
thyristors 12 and 13 is late in the cycle. It will be noted that
the dash-dot line which represents the point of conduction of
thyristors 15 and 16 goes right off the graph. These thyristors do
not conduct under this condition, and only thyristors connected to
the lower potential taps are fired. The various conductive
conditions are illustrated in FIG. 2
As the load resistance changes as shown at B, current errors occur
which when integrated by amplifier 28 and amplified by amplifiers
such as 29, produce reference represented by the broken lines on C.
The dotted line represents the reference used to control thyristors
12 and 13 while the dash-dot line represents the reference used to
control thyristors 15 and 16.
In time period t.sub.1 the dotted line cuts the ramp early in the
cycle and causes thyristors 12 and 13 to fire. Later in the cycle
the dash-dot line also cuts the ramp and thyristors 15 and 16 are
fired. The resultant voltage and current waveforms are shown at D
and E respectively and the pulse trains used to fire thyristor 15
for example are shown at C'.
In time period t.sub.2 the load resistance increases and more
voltage is required to produce the same current. Thyristors 12 and
13 fire for the whole cycle while thyristors 15 and 16 fire for a
substantial portion of the cycle.
In time period t.sub.4 the load resistance is zero (with contacts
20d and 20c closed). The error signal is such that the amplifier
associated with thyristors 15 and 16 is inoperative (due to its
bias) and only thyristors 12 and 13 are fired. Since these are
feeding a reactive load they conduct through the zero transistion.
The series of pulses as shown at C' or a prolonged firing signal is
used to ensure thyristors turn on in this period.
It should be evident that this system as illustrated is only one
specific example of our invention. Only two thyristor pairs have
been shown on transformer 10, but it is evident that a plurality of
thyristor pairs could be included and connected to various
transformer taps depending upon the degree of control required. The
more thyristor pairs provided the less waveform distortion will be
produced. It will also be seen that load transformer 14 is
necessary in the present situation only because of the voltages and
currents being controlled. Recognizing the voltage limitations of
the thyristors, it was necessary that they be operated at a
restricted voltage rating while the load was operated at a higher
voltage. If this were not the case, transformer 14 could have been
omitted. It will also be understood that the specific circuitry for
producing the pulses, while preferred, is not the only possible
arrangement.
It is also noted that the square law device designated 21 which
produces a signal proportional to the square of the current in
current transformer 19 may take various forms. For example, it may
be produced by a proper arrangement of a Hall generator. Integrated
devices are available however which will produce a square law
signal in response to an input signal.
Finally, it is noted that the whole apparatus is assumed to be
operated on a 60-cycle alternating supply, and while it is
necessary that the various supplies all be from the same source,
there is no reason why it should not operate at some other
frequency dependent on the available source of supply.
* * * * *