JPS5875794A - Illumination system - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to electrical lighting systems, and more particularly to lighting systems with substantially increased efficiency.

The present invention is particularly directed to systems of the type using metal halides, mercury vapor, high or low pressure sodium, or fluorescent lamps and lamp ballasts.

All of the aforementioned lamps and systems using them have been improved with the primary goal of using power in an all-encompassing manner. As the price of energy has increased, consideration of the need for practical, expensive, and efficient systems that conserve energy has become even more important, while increasing user efficiency remains generally unsatisfactory. It has become.

Those skilled in the art are very familiar with the various ballast and system configurations. Accordingly, ballasts and systems in their various forms have been described in the literature. Electric ballasts have been developed in recent years, particularly for the purpose of increasing the efficiency of lighting systems and to further extend the average life of lamps beyond those experienced with electromagnetic ballasts in the past.

Further background art and examples are provided in U.S. Pat. No. 4,277,726.
It is given by the number. This U.S. patent was filed by Robert Bui.
Robert V. Burke, July 7, 1981, and assigned to the applicant. The technical content of that US patent is quoted here.

As described in U.S. Pat. No. 4,277,726, electronic ballasts typically include a rectifier, or transistor, that converts an AC line voltage to a DC voltage. This transistor is placed in an oscillator circuit, and this transistor and a transformer work together to transform the AC current generated on the primary side of the transformer by the action of the oscillator, and generate a high voltage on the secondary side of the transformer. , providing full power to the lamp.

We believe that Hand 1 provides a basic starting point from the teachings and direction of the art by providing lighting systems. On the other hand, although polyphase power has been used to power motors and other heavy-duty electrical equipment, Applicants have discovered that the polyphase lighting system of the present invention provides an unexpectedly fast payback period (pa
y bac , k period ) f and to cancel out certain ballast components.

By way of example, a preferred polyphase ballast will be described. This ballast is suitable for use in the system described above. Essentially, the preferred ballast is a modification of that described in the above-cited US Pat. No. 4,277,726.

As will be apparent from the description of the preferred embodiments, the system of the present invention utilizes certain power-handling components.
ndling components). Therefore, the result is cost savings, increased reliability, and increased efficiency.

More particularly, the present invention eliminates the filters commonly used in the rectification stage of electronic ballasts. The unexpected result is that the omission of this filter substantially increases the overall cost efficiency of the lighting system. Other prior art ballasts were similarly modified in accordance with the present invention and are used as a basis for comparison.

Referring first to FIG. 1, there is shown a ballast described in U.S. Pat. No. 4,277,726. This ballast includes full-wavelength diodes 21, 22, 23. Shown as including a rectifier.

It is known in the art that a full wave rectified DC voltage before filtering has a superimposed ripple voltage of magnitude 48 at a frequency twice the power supply frequency, for example 2 x 60 H2. (Reference data for radio technology J (Reference Da
te ForRadio Engineers) 4th
ed., ITT% NY; 1964, p. 306. ) The configuration and operation of the circuit shown in FIG. 1 will be described in detail below. The circuit includes a first NPN transistor 1, a second NPN transistor 3, and a transformer 5. The transformer 5 has a centrally tapped primary winding 7, a low voltage feedback winding 9, and a centrally tapped high voltage secondary winding a1.
1, comprising low-voltage secondary windings 13, 15 and 17 and an additional low-voltage winding 19, all windings of which are wound on a magnetic ferrite core with a gap indicated by two parallel lines. ing. The collector of transistor 1 is connected to one terminal of the primary winding, and the collector of transistor 3 is connected to the other terminal of primary winding 7. capacitor 4 is 1
The primary winding 7 and the capacitor 4 are connected between both terminals of the secondary winding, thereby forming an L-C valve train resonant circuit. For purposes of illustration, lead wire 6 is the neutral or common reference point of the circuit. Wire 6 is connected to the emitters of transistors 1 and 3, and Zener diode 8 is connected to primary winding 7.
is connected between the center tap of and the emitters of both transistors 1 and 3. Diode 8 is polarized such that its cathode is connected to the center tap of the primary winding.

This will be explained in more detail below. Inductor 1° has one terminal connected to the center tap of primary winding T, so that inductor 10 and primary winding 7 are in series. Both terminals of feedback winding 9 are connected to the respective bases of transistors 1 and 3 as shown. The circuit described above is generally a push-pull type inverter oscillator circuit. On the left-hand side of the figure, terminals are provided for connecting a 120V AC power source to arms a and b of a conventional bridge rectifier. The bridge rectifier consists of four rectifier diodes 21°22.23. and 2
The bridge rectifier and other circuit elements are protected by a varistor 25 connected between both incoming arms of the bridge. Resettable thermal circuit breaker 29
is connected to the first input arm of the illustrated 120V AC line source and bridge rectifier in series with one lead and an electrical ON-OFF light switch. The other power lead is connected directly to the other input arm of the bridge. One output arm d of the bridge rectifier is connected to the neutral wire 6 in series with a choke inductor 34, and the other output arm C is connected to one terminal of the inductor, 31. Capacitor 32 is connected to inductors 31 and 3
Connected to one terminal of 4.

A capacitor 33 having a larger capacitance value than the capacitor 32 is connected between the output terminals of the chokes 31 and 34. The output terminal of inductor 31 is connected to the input terminal of inductor 10.

Inductor 34 has the same inductance value as inductor 31. As shown by the U-shaped line indicating the iron core, both inductors 31 and 34 are formed in one or two turns on the same iron core, which is juxtaposed. This is for practical and economical construction. This is because the current flowing through each winding during operation flows in opposite directions.

Obviously, the bridge rectifier and elements 31, 32 and 33
A "π" filter consisting of provides a partially filtered DC voltage and current from a 120V AC RMS input.

A high resistance 35 is connected between the output terminal of the inductor 31 and the base of the transistor 1. Resistor 35 is also 2
The secondary winding 9 is connected to the base of the transistor 3.

The low voltage secondary winding 19 has a diode 26 and a capacitor 2.
8 in series, and the cathode terminal of the diode 26 is connected to the positive terminal of the capacitor indicated by "+". One terminal of the capacitor 28 and the winding 19 are both connected to the neutral point 6.

A typical Darlington transistor 30 consists of two transistors, each of the NPN type.

In this Darlington transistor, the collectors of each transistor are connected in common, the emitter of the first transistor is connected to the base of the second transistor, and the emitter of the second transistor provides the output shown symbolically in the figure. The collector f of the Darlington transistor is connected to the "+" terminal of the capacitor 28. The emitter g of the Darlington transistor is connected in series with a low resistance 39, which is connected to the base of the transistor 1, and connected to the base of the transistor 3 via the winding 9. A medium resistor 36 is connected between the emitter and collector terminals of Darlington transistor 30 to provide a Schundt resistance path of Darlington transistor 30 to resistor 39. A resistor 37 and a capacitor 38 are connected in series in a circuit between the output terminal of the inductor 31 and the neutral circuit 6, forming an R-C type timing network. Zener diode 40 is connected between the junction between elements 37 and 38 and the control terminal of Darlington transistor 30. Referring to the right side of FIG. 1, a low voltage secondary winding 13 is connected across the terminals of the first fluorescent lamp 41 heater, and a low voltage secondary winding 15 is connected across the heater of the second fluorescent lamp 43. connected between both terminals. A low, voltage secondary winding 17 is connected to each of the remaining heater terminals of lamps 41 and 43 in the parallel circuit. As shown, the center tap of the high voltage secondary winding 11 is grounded for protection purposes, and one terminal of the secondary winding 11 is connected to a lamp 43 in series with a series load capacitor 45.
connected to the heater terminal.

The connection between the second capacitor 47 and the lamp is the capacitor 4
5 between one terminal of the winding 11 and the corresponding heater cathode terminal of the lamp 41. The other terminal of the secondary winding 11 is commonly connected to a lead emerging from the low voltage winding 17. Winding 11 is also applied to the remaining cathode of each lamp as it appears in the circuit, connecting the series combination of the capacitor and lamp to the secondary winding.

The operation of the ballast shown in the first figure will be explained. The user operates the light switch 114 to select 120V50 or 60V.
Hz AC line voltage through diodes 21 to 24□
connected to a bridge rectifier consisting of Current flows into the bridge via thermal breaker 29. The output appears between the "+" and "-" output arms of the bridge rectifier as a full-wave rectified pulsating DC voltage.

This rectified voltage is applied to the inductor 31 and capacitor 32.3.
A π filter consisting of 3 is supplied. This π filter partially "smoothes" or filters the rippled DC to provide a DC voltage with a moderate ripple component. This acts as a ballast DC voltage source, as known to those skilled in the art, and provides a partially filtered DC voltage even under a "load" that extracts current. Generally, direct current flows into the circuit from the "+" side of the power supply and returns from the circuit via the neutral point 6. Inductor 31 and inductor 34 have the additional function of providing high electrical impedance to high frequency energy. The radio frequency energy is substantially above the line frequency of 50 or 60 Hz. Inductors 31 and 34 also prevent transients occurring in the circuit from reaching the bridge rectifier circuit or the AC line and improve the input power factor. Preferably, a small DC current flows from the positive side power λ of the DC power supply through the high resistance 35 to the bases of the transistors 1 and 3. This current is sufficient to provide the minimum base drive current required by transistors 1 and 3, and this current causes transistors 1 and 3 to operate in oscillation mode. In this oscillation mode, transistor 1
and 3 conduct current through each half of the primary winding 7 at all possible ambient temperatures at which the ballast may be used.

Other known means for initiating oscillation may be used if desired under particular operating conditions. When one of the transistors starts operating, a direct current flows through the inductor 10 to the center tap of the primary winding. Transistor 1 is now biased into conduction or an ON state, allowing current to flow through half of the primary winding and the collector of transistor 1 to neutral point 6 and back to the negative power source terminal and into the transformer iron core. generates magnetic flux lines corresponding to and induces a voltage in the secondary winding.

When transistor 3 is conducting and transistor 1 is non-conducting, current flows through inductor 10, the center tap of the primary winding, the other half of the primary winding, and the collector, collector of transistor 3, and It returns again to the power source via neutral point 6. This generates a magnetic flux of opposite polarity corresponding to the transformer core. This is because the current flowing through the other half's primary winding is in the opposite direction. Since the currents flowing through the primary windings are in opposite directions, the operation in which all the currents flow alternately through the primary windings is
The magnetic action generates a changing magnetic field within the transformer core. The AC voltage is the secondary winding 11.13.15.9
．． 17. and 19.

Transistor 1.371- by feedback winding
The voltage applied to the base of transistor 4 is 180° out of phase with the voltage applied to the other transistor of transistor 3. This is because each base is connected to the opposite terminal of the secondary winding 9, 1□
1 has been done. This feedback voltage biases one transistor into a conducting or ON state and biases the other transistor into an OFF state.

When the current in the primary winding flows in the opposite direction, the voltage in the secondary winding 9 also has the opposite polarity, thus making the other transistor iON and one transistor j
z O, FF Iasu to put it in F state. The speed or frequency at which transistors 1 and 3 alternately switch ON and OFF states in self-holding oscillation due to voltage feedback is the oscillator frequency, which is governed by the inherent electrical characteristics of the circuit. The effective inductance of the transformer introduced in the primary winding, the inductance imparted to the transistor circuit in the primary winding by the electrical load and the electrical load including the capacitor 4, the capacitance, and the resistance etc. Electrical characteristics are essentially parallel resonance L
C circuit is defined, and this LCl path switches at its resonant frequency to produce a high frequency oscillation essentially at that resonant frequency.

In this inductor circuit, the inductance of the inductor 10 is larger than the self-inductance of the primary winding 7. That is, the ratio between the inductor 10 and the self-inductance of the primary winding 7 is at least 2.5:1 and at most 10:1 to 100:1. Due to this relationship, 1
The current flowing into the next winding will be essentially constant.

As mentioned above, the secondary winding 1 generated by the transformer operation
The low AC voltage of 9 is half-wave rectified by a charging capacitor 28 to a rectifier 26 arranged in a half-wave rectifying filter circuit.

Thereby, a DC voltage is generated across the capacitor 28, which is marked with a positive polarity. Current from this DC source first flows through medium sized resistor 36 and low resistor 39 to the bases of transistors 1 and 3, thereby providing additional operating base drive current to those transistors. Therefore, the very small bias current supplied via the resistor 35 mentioned above is supplemented. By supplying the full drive current instead of directly from the bridge rectifier in this manner, I2R losses and the overall electrical efficiency of the unit are improved.

Please pay attention to the following. That is, when a single-stage fluorescent lamp operates as described below, the fluorescent lamp changes from an unprecedented capacitance in the load to an effective resistance in the secondary winding. This effect is then reflected back to the primary winding as a separate load on the resonant circuit.

This results in new circuit conditions resulting in a new resonant frequency for the inverter oscillator.

The following points should be noted. That is, in addition to this function as a resonant circuit element, the capacitor 4 serves to absorb any high frequency energy that may arise as a result of the electrical nonlinearity of the electrical load. This electrical load is otherwise coupled back to the primary winding by the leakage inductance from the primary to the secondary winding of the transformer 5. The capacitor 4 also smoothes out any transient phenomena that may exist from the inductor 10 to that circuit location. The Zugner diode 8 allows current to flow in one direction from the anode to the cathode like a normal diode, and in the opposite direction, the voltage level in the opposite direction is the characteristic of the diode. The diode prevents current from flowing in the reverse direction until the breakdown voltage level is exceeded, at which point the diode allows current to flow in the reverse direction. Zener diodes are normally non-conducting and are polarized so that they conduct current in the opposite direction only when a sufficiently large transient voltage occurs across their terminals.

This diode blocks all such voltage transients from reaching the transistor.

The alternating current voltage generated in the secondary winding, especially the high voltage in the secondary winding 11, is partially affected by the voltage in the primary winding 7 in the same way as the turns ratio N, that is, the turns ratio of the secondary winding to the primary winding. depends on By providing only a limited base drive current to the inverter transistors through resistor 36, the primary winding voltage and primary winding current are necessarily limited, which in turn limits the smaller magnetic field. Therefore, a smaller voltage is applied to the secondary winding 11 than when the transistors 1 and 3 are fully conductive, that is, when they are saturated. If the alternating voltage applied to windings 13, 15 and 17 is in any case very low, on the order of 4v, the effect of this limitation on the purpose of those windings is unimportant, and the current at that low voltage is Each heater winding supplies an associated lamp heater. However, capacitor 4
5 and lamp 43, and capacitor 47 and lamp 4
The voltages applied to each of the series load circuits, each constructed from 1, are such that the voltage developed at a given time in the secondary winding 11 is less than the required starting voltage of those lamps. This voltage is also about 0% smaller than the starting voltage. Therefore, the lower peak of the AC voltage applied to the winding 11,
That is, a peak voltage below the voltage at which the lamp starts is important in this portion of the circuit in that the lamp will not start even if the lamp cathode is heated.

Again, for the DC output at the right terminal of inductor 31 shown in the first diagram, when 120V AC is applied to the circuit and rectified, the DC current flows through resistor 37 and capacitor 38, and the timing circuitry. As will be understood by those skilled in the art, a series RC circuit has the property that a predetermined DC voltage φ; when applied to that RC circuit, current flows through the resistor and slowly charges the capacitor over a period of time. . This is because a voltage is generated across the capacitor. As a result, the voltage level of the power supply becomes uniform. The voltage generated across the capacitor 38 varies logarithmically with respect to time, as is well known. The time required for the voltage across the capacitor to equalize to a voltage level of approximately 63% of the voltage level of the applied DC source is equal to (1/RC) seconds. Here, R is a resistance value in ohms, and C is a capacitance value in farads. However, since the voltage appearing at the terminals of inductor 31 is not constant and contains some ripple as mentioned above, the time constant of the circuit is approximately equal to the value determined mathematically from the above expression, but the exact value of the circuit is The time constant is most preferably determined by experiment.

At some time after applying voltage to the ballast, capacitor 38
reaches a first predetermined voltage level, approximately 20V.

Applicant regards this voltage as the "trigger voltage level" for the full Zener diode 40. And this is 1
Approximately 0 from the time 20V AC is applied to the ballast input
．． Happens after 15 seconds. At the same time, this means that the secondary winding 1
1 allows the heater winding to supply heater current to the lamp cathode for an insufficient time to start the heater lamp.

During that period of time, the voltage on capacitor 38 is
1 breakdown of Zener diode 40 or commonly used condition, i.e. Darlington transistor 3
When the control input is 0, a "trigger voltage level" is reached that switches the state to flow a reverse direct current in the direction of the base h. Darlington transistor 30 then begins conducting current from capacitor 28 of the auxiliary DC power source through the collector and emitter of Darlington transistor 30.

This additional current flows in the shunt of resistor 36 and further through resistor 39 to the bases of switching transistors 1 and 3.

Considering the aforementioned mode of operation, as the base drive current increases, each switching transistor conducts a large current in each half cycle of the high frequency alternating current in the aforementioned resonant mode of operation. At this time, a larger current flows in the primary winding of the transformer, a large magnetic field is generated in the magnetic core, and finally the high voltage secondary winding,... Output 2 of the device
A large voltage is induced in each of the secondary windings.

Paying attention to the capacitor 38 again, as a result of the continuous charging of the capacitor 38, the voltage between both terminals of the capacitor 38 rises above the threshold voltage, and the Darlington transistor 30 at the base h receives a large drive. Further sufficient conduction is provided from the emitter to the collector circuit. Therefore, a larger base drive current is supplied to the bases of switching transistors 1 and 3. This current continues to increase until the time when transistor 30 reaches saturation current. This current saturation has the effect of providing an upper limit to the base drive current applied to the bases of transistors 1 and 3 via Darlington transistor 30.

As previously mentioned, the high voltage, high frequency alternating current originally generated in winding 11 has a first level, which level is less than the voltage that starts the lamp. The voltage of the high voltage secondary winding 11 is 1
The level gradually rises to the lamp starting voltage required by the two lamps. The lamp current gradually begins to give a more complete desired brightness.

Because no two fluorescent lamps are exactly the same, one lamp requires a lower voltage to start than the other, and one lamp requires approximately 100% less voltage than the other lamp.
/Starts in 10 seconds less time.

Therefore, in the illustrated device, Darlington transistor 30
provides the first function of the electronic switch. This switch switches from a non-conducting state to a conducting state after said predetermined time and provides base drive to the switching transistor 3 of the inverter oscillator part of the circuit starting from a first level set primarily by the resistor 36. Increase current. Darlington transistor 30 then functions as a control amplifier that gradually increases the base drive level to a higher second level.

Before activation, each lamp functions as a small capacitance within the system. however,
- When started, the lamp draws a large current and consumes energy. Here, the series capacitance serves to limit the current flowing in the circuit. As a result, the resonant frequency of the inverter oscillator shifts to a lower frequency under high current loads.

As Darlington transistor 30 continues to increase the base drive current, the voltage on the secondary winding increases further until the second lamp starts.

Next, when the second lamp is activated, the current flowing through the secondary winding 11 further increases. As a result, the speed or frequency of the oscillator is further reduced. The applicant considers this mode of operation to be "soft start" in two senses:
Avoid starting both lamps at the same time, which would create large surge currents in the secondary winding circuits that could create hum-inducing transients, and avoid starting both lamps at the same time, since the output voltage will gradually rise to the starting level. This means that the starting current changes more slowly than in other cases. In my opinion, the latter result prevents stress on the lamp cathode and extends lamp life.

Please refer to Figure 2. Illustrated in this figure is a ballast 100 of a preferred embodiment of the present invention. Ballast 10
0 is the line φA of the AC power source of 60 Hz, respectively.
, φB, φC, thereby lighting a pair of air discharge lamps 41,43'r. In particular, the ballast 100 includes diodes 102, 104, 106, 108,
It includes a three-phase full-wave rectifier consisting of 110 and 112.

This rectifier produces a DC voltage between lines 114, 116 with a ripple of 4.2% at 6 times the frequency of the power supply, or 360H2. (See page 3, 07 of ``Reference Data for Radio Engineers.'') By substantially reducing the ripple voltage, the capacitors shown in the first diagram: the capacitors 32, 33 and the inductor 31, etc. A rectifier can be omitted. Accordingly, the unwavering rectified current shown in FIG. 2 is coupled directly to the connection between inductor 110 and resistor 135.

What should be noted is that the ordinary power switch 1 shown in the first diagram
14 (discovered by the applicant) Square D
It may also be preferably replaced by a three-phase power switch, such as a Class 2510, type manual motor start switch manufactured by 5quare D Company. The switch may be physically housed in the space typically reserved for interior light switches.

The three-phase AC power sources φA, φB, φC and the diode bridge circuits 102, 104゜106, 108, 110
, 112 is almost the same as that shown in the first figure.

The large inductors 31, 34 (FIG. 1) are not needed in the circuit shown in FIG. If so, those inductors 31
．． 34 is offset by the high power factor of the three-phase power. a pair of 1
A mH inductor is coupled between the rectifier and the inverter. However, these inductors are provided for a different purpose than that of the circuit of FIG. That is, the purpose is to prevent the inverter frequency from being reflected onto the power line.

Darlington transistor 30 (FIG. 1) and associated Zener diode 40 have been omitted from FIG. Darlington transistor 30 is used in the single phase circuit of FIG. 1 to prevent two lamps from starting at the same time. Such simultaneous lighting is achieved by the Darlington transistor 30 and the Zener diode 40.
This occurs due to the dimensions of the capacitor 31 (FIG. 1) if it is not provided. In particular, the open circuit voltage of capacitor 33 needs to be large to provide the required load voltage of that capacitor 33. The large open circuit voltage instantly activates the two lamps.

In the circuit shown in the second diagram, since the adjustment between the no-load voltage and the maximum load voltage of the capacitor 33 is inherently good, there is no need to provide the Darlington transistor or the like.

The lighting systems and ballasts of the present invention have many advantages over conventional systems and devices. 1st
Second, the elimination of rectifying filters directly saves money and improves ballast reliability.
And the efficiency of the ballast can be increased.

To save manufacturing costs, two large capacitors 3
This is achieved by omitting 2.33 and the inductor 31 (first shown).

The addition of two diodes to form a three-phase rectifier does not significantly impede the achievement of cost savings.

The reason for the increased reliability of the ballast is that the omitted capacitor is generally an electrolytic capacitor, which is known to those skilled in the art as a statistical weak link in reliability calculations. It is being Even if you wanted to provide some type of RFRJ filtering to prevent the oscillator frequency from being reflected through the AC power line, the capacitor and The inductor will be extremely small and inexpensive. Even so, it is possible to further omit the step of applying a coating agent to the inductor.

This step is performed in the known prior art to eliminate the 60 Hz hum.

Ballast efficiency is improved because typical solid ballast rectifier filters can suffer losses of up to 5%.

The inductor in such a filter causes both so-called "2R" loss due to the inherent electrical resistance of the inductor and electromagnetic induction loss caused by the magnetic flux in the iron core of the inductor. The electrolytic capacitors commonly used in such filters account for approximately 1 out of the 5 losses noted above. T'riad/Vtra
It has been discovered that the ballast of the present invention achieves efficiencies of 94-95%, while conventional electronic ballasts such as part numbers B140R8 and 8240R8 achieve efficiencies as low as 90-.

By reducing these losses, the operating temperature of the ballast is reduced. Since reliability is inversely proportional to operating temperature, lowering the operating temperature increases reliability.

FIGS. 3-5 illustrate the relative efficiency, power loss, and power factor achieved by the ballast of the system of the present invention compared to that of currently commercially available single phase electric ballasts. In particular, a prototype ballast of the present invention constructed in accordance with the preferred embodiment shown in FIG. 2 was connected to a number of test apparatuses shown in FIG.

Triad/Utrad of Hunting
on, Part number B1 sold by Indian
A 120v single phase electric ballast designated 40R8 was installed in the sixth
Connected to the test equipment shown. Similarly, Tr, iad
/Part number B240R8 270 sold by Utrad
■The electric ballast was connected to the test equipment shown in Figure 6.

As shown in FIGS. 3-5, the ballast of the present invention was significantly more efficient than single phase ballasts, had lower power losses, and provided a more ideal power factor.

With the system of the present invention, line loads are also essentially balanced. Those skilled in the art will appreciate that three-phase power is typically routed within non-residential buildings and that the phases are separated into three branches;
It is known that each school independently powers a number of lamps. Although it is desirable to have equal loading on each branch, since the banks of lights in each branch operate independently, it is common for almost all "ON" and rOF
This results in an unbalanced load on the combination of FJ ballasts. In contrast, the system of the present invention is essentially balanced for all ballast "ON" and rOFFJ combinations.

Finally, the high efficiency and balanced load characteristics of the ballast of the present invention - more lamps per circuit, thereby reducing the amount of wiring required.

The above description is, of course, only an example, and the scope of the present invention should not be limited to the example.

The invention is limited only by the scope of the claims.

It will be appreciated by those skilled in the art that many changes and modifications can be made without departing from the scope of the invention.

[Brief explanation of the drawing]

FIG. 1 is a diagram illustrating the ballast described in U.S. Pat. No. 4,277,726, FIG. 2 is a diagram illustrating a preferred lighting system constructed in accordance with the present invention, and FIGS. 6 and 7 are graphs of test data showing the efficiency, power loss, and power factor, respectively, of a lighting system constructed in accordance with the present invention; FIGS. The first
FIG. 3 is a diagram showing test circuits used to obtain data for the single-phase ballast and the three-phase ballast shown in FIG. 3 and FIG. 2, respectively. [Explanation of symbols of main parts] Air discharge lamp・・・・・・・・・141,143 Multiphase power source・・・・・・・・・φA, φB,
φC ballast・・・・・・・・・・・・・・・・・・1
00 Applicant: Ridon Systems, Inc. Engraving of drawings (no changes in content) 432- Output power (W) Output power (W) Human power (W) F19J Procedural amendment November 22, 1980 Commissioner of the Patent Office Mr. Kazuo Wakasugi 1, Indication of the case 1982 Patent Application No. 178237 2, Title of the invention 3, Person making the amendment Relationship to the case Patent applicant 4, Agent 5, Subject of the amendment "Drawings" (1) We will submit one official drawing as shown in the attached sheet. 111

Claims (1)

Claims: 1. For use in lighting systems of the type that include air discharge lamps that require a starting voltage and an operating voltage, the lamp being coupled to a multiphase power source that provides the starting and operating voltages. A ballast featuring 2. The ballast according to claim 1, wherein the multiphase power source is a three-phase power source. 3. a multiphase AC power source; a plurality of air discharge lamps of a type requiring a starting voltage and an operating voltage; and coupling the lamps to the multiphase power source and applying the starting voltage until the lamps start. A lighting system comprising: a ballast operative to provide an operating voltage after the lamp has been activated. 4. A method for operating a complete lighting system of the type including an air discharge lamp requiring a starting voltage and an operating voltage, comprising the step of coupling a multiphase alternating current power source to said system. How to do it.