JPH0992478A - Discharge lamp lighting device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To make easy start-up of discharge lamps by controlling impedance controlled variables according to whether all the discharge lamps are turned on or off or some of them are turned on but others are not, and applying appropriate voltages to them. SOLUTION: When two fluorescent lamps 66, 69 are turned on, voltage regulation diodes 76, 78 are put into conduction to connect resistances 87, 88 in parallel with a resistance 86, and control impedance is controlled to control the oscillating frequency of an inverter circuit so as to apply an appropriate lamp voltage VL1 to each of the fluorescent lamps. When one of the lamps is turned on, one of the voltage regulation diodes is put into conduction to connect the resistance 87 or 88 in parallel with the resistance 86, and the control impedance is enhanced to apply a proper lamp voltage VL2 to the fluorescent lamp that is off. Further when the two lamps are turned on, only the resistance 86 is used to further enhance the control impedance to apply a lamp voltage VL3 (<VL1 ) appropriate for keeping the fluorescent lamps on.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a discharge lamp lighting device for lighting a plurality of discharge lamps such as fluorescent lamps in parallel by using an inverter circuit for controlling self-excited oscillation.

[0002]

2. Description of the Related Art In this type of discharge lamp lighting device, an inverter circuit 2 is connected to a DC power source 1 as shown in FIG. The DC power supply 1 is composed of, for example, a circuit that performs full-wave rectification and smoothing of an AC power supply, and the negative electrode side is grounded. The inverter circuit 2 includes a pair of MOS type FETs for the DC power supply 1.
A series circuit of (field effect transistors) 3 and 4 is connected in parallel, and a series circuit of resistors 5 and 6 and capacitors 7 and 8 is connected between the gates and sources of the FETs 3 and 4, respectively. The connection point of the FETs 3 and 4 is connected to one end of the first current transformer 9, and the series circuit of the resistor 5 and the capacitor 7 is magnetically coupled to the first current transformer 9 via the resistor 10. 2 current transformer 11
Are connected in parallel, and a third current transformer 13 magnetically coupled to the first current transformer 9 is connected in parallel to the series circuit of the resistor 6 and the capacitor 8 via the resistor 12.

The other end of the first current transformer 9 is
The capacitor 14 and the first choke coil 15 are connected in series to one end of one filament electrode 16a of the first fluorescent lamp 16, and the capacitor 17 and the second choke coil 18 are connected in series to form a second coil electrode. It is connected to one end of one filament electrode 19a of the fluorescent lamp 19. One end of the other filament electrode 16b, 19b of each fluorescent lamp 16, 19 is connected to the source of the FET 4. The capacitor 20 is placed between the other ends of the filament electrodes 16a and 16b of the fluorescent lamp 16.
And each filament electrode 1 of the fluorescent lamp 19
A capacitor 21 is connected between the other ends of 9a and 19b.

A series voltage dividing circuit of capacitors 22 and 23 is connected between one ends of the filament electrodes 16a and 16b of the fluorescent lamp 16, and a capacitor 2 is connected between one ends of the filament electrodes 19a and 19b of the fluorescent lamp 19.
4 and 25 series voltage dividers are connected to connect capacitors 22 and 2
The voltage dividing point of 3 is connected to one input terminal of the 2-input AND gate circuit 27 via the constant voltage diode 26, and the voltage dividing point of the capacitors 24 and 25 is connected to the AND gate circuit 27 via the constant voltage diode 28. It is connected to the other input terminal. The output terminal of the AND gate circuit 27 is connected to the base of the NPN transistor 29.

Further, a resistor 30, between the + VE terminal and the ground,
The emitter and collector of the PNP type transistor 31, the series circuit of the resistor 32 and the resistor 33 are connected, and the resistor 3 is connected.
A series circuit of 4, 35 and a resistor 36 is connected. The base of the transistor 31 is connected to the connection point between the resistors 34 and 35. The collector and the emitter of the transistor 29 are connected in parallel to the resistor 36.

Further, a fourth current transformer 37 magnetically coupled to the first current transformer 9 is provided, and one end of the fourth current transformer 37 is connected to the cathode of the diode 39 and the anode of the diode 40 via the capacitor 38. The other end is connected to the anode of the diode 39 and is grounded. A capacitor 41 is connected in parallel to the diode 39 via the diode 40. Then, a MOS type F is added to the capacitor 41.
Connect the drain and source of ET42 in parallel,
The gate of T42 is connected to the connection point of the resistors 32 and 33.

This device comprises a series resonance circuit of a capacitor 14, a first choke coil 15 and a capacitor 20,
Resonance is maintained by alternately switching the FETs 3 and 4 at a frequency close to the resonance frequency with respect to the resonance system of the series resonance circuit of the capacitor 17, the second choke coil 18, and the capacitor 21. When the inverter circuit 2 operates, an alternating current flows through the first current transformer 9, and a part of this current also flows through the magnetically coupled second to fourth current transformers 11, 13, 37. Since each of the current transformers 9, 11, 13, 37 utilizes supersaturated magnetism, the inverter circuit 2 can be adjusted by adjusting the saturation amount.
The oscillation frequency of can be controlled.

That is, the voltage applied to the fluorescent lamp 16 is divided by the capacitors 22 and 23, and when the voltage at this dividing point exceeds the Zener voltage of the constant voltage diode 26, the constant voltage diode 26 becomes conductive, and the fluorescent light is emitted. The voltage applied to the lamp 19 is divided by the capacitors 24 and 25, and when the voltage at this voltage dividing point exceeds the Zener voltage of the constant voltage diode 28, the constant voltage diode 28 becomes conductive. Then, when both of the constant voltage diodes 26 and 28 become conductive, the output level of the AND gate circuit 27 becomes high level,
The transistor 29 is turned on.

When the transistor 29 is turned on, the resistance 3
Since 6 is short-circuited, the base potential of the transistor 31 decreases and the base current of the transistor 31 increases. As a result, the impedance between the emitter and collector of the transistor 31 decreases, and the gate potential of the FET 42 increases. In this way, the impedance between the drain and source of the FET 42 decreases, the charge of the capacitor 41 is extracted, and the impedance of the fourth current transformer 37 decreases. As a result, the first current transformer 9
Magnetic saturation is weakened, and as a result, control is performed in the direction of lowering the oscillation frequency of the inverter circuit 2.

When the fluorescent lamps 16 and 19 are not turned on at the time of starting, the impedance control amount is determined so that a predetermined tube voltage VL appropriate for starting the lighting of the fluorescent lamps can be applied. The graph of FIG. 7 shows the frequency-tube voltage characteristic g1 of the non-lighting resonance mode and the frequency-tube voltage characteristic g2 of the lightning resonance mode. The proper tube voltage at the non-lighting is VL1. The oscillation frequency is f (pressurized).

The graph of FIG. 8 shows the impedance control amount-oscillation frequency characteristic g3 when both fluorescent lamps 16 and 19 are not lit, and the impedance control amount-oscillation frequency characteristic when one lamp is lit and one lamp is not lit. g4 shows the impedance control amount-oscillation frequency characteristic g5 when both lamps are lit, but the impedance control amount-oscillation frequency characteristic g3
Since the oscillating frequency when both lamps are not lit is f (pressurized), the impedance control amount is Z4.

The oscillation frequency when both lamps are lit is f (lit), and the impedance control amount at this time is Z2 from the impedance control amount-oscillation frequency characteristic g5. By the way, fluorescent lamps have variations in characteristics, and a plurality of fluorescent lamps are not always lit at the same time. In the circuit of FIG. 6, if the fluorescent lamp 16 is turned on and the fluorescent lamp 19 is not turned on,
Although the voltage at the voltage dividing point of the capacitors 22 and 23 decreases, the voltage at the voltage dividing point of the capacitors 24 and 25 does not decrease. Therefore, the constant voltage diode 26 becomes non-conductive, but the constant voltage diode 28 remains conductive. However, when the constant voltage diode 26 becomes non-conductive, the output of the AND gate circuit 27 becomes low level, and the transistor 29 is turned off.

As a result, the short circuit of the resistor 36 is released, the base potential of the transistor 31 rises, and the base current of the transistor 31 decreases. As a result, the impedance between the emitter and collector of the transistor 31 increases, and the gate potential of the FET 42 decreases. Thus
The impedance between the drain and the source of the FET 42 increases, and the impedance of the fourth current transformer 37 increases. As a result, the magnetic saturation of the first current transformer 9 is increased, and as a result, control is performed in the direction of increasing the oscillation frequency of the inverter circuit 2.

That is, the impedance control amount at this time is Z2 as in the case where both lamps are turned on. Therefore, from the impedance control amount-oscillation frequency characteristic g4 in FIG. F (AND) higher than (pressurized). Fluorescent lamp 19 that is not lit at this time
The tube voltage VL applied to the circuit is the frequency-tube voltage characteristic g in FIG.
The voltage changes from 1 to VL10, which is lower than the originally required tube voltage VL1.

Further, another discharge lamp lighting device, as shown in FIG. 9, is a constant voltage diode 26, 28 of the conventional example described above.
And instead of the AND gate circuit 27, a diode 43,
The wired OR circuit 44 and the constant voltage diode 45 are used.
If either of the voltage dividing points of the capacitors 24 and 25 is high, the constant voltage diode 45 is turned on and the transistor 29 is turned on.
Is the one that works on.

The graph of FIG. 10 is similar to the graph of FIG.
The frequency-tube voltage characteristic g1 of the non-lighting resonance mode and the frequency-tube voltage characteristic g2 of the lightning resonance mode are shown.
Further, the graph of FIG. 11 is similar to the graph of FIG. 8, the impedance control amount when both the fluorescent lamps 16 and 19 are not lit-oscillation frequency characteristic g3, the impedance when one lamp is lit, and the impedance when one lamp is not lit. Control amount-oscillation frequency characteristic g4,
The figure shows the impedance control amount-oscillation frequency characteristic g5 when both lights are on.

In this device, when one lamp is turned on and one lamp is not turned on, the transistor 29 is turned on, so that the resistor 36 is short-circuited and the base potential of the transistor 31 is lowered. The base current of increases. As a result, the impedance between the emitter and collector of the transistor 31 is lowered, and the gate potential of the FET 42 is raised, so that the impedance between the drain and source of the FET 42 is lowered and the charge of the capacitor 41 is drawn out. Lower the impedance of. As a result, the magnetic saturation of the first current transformer 9 becomes weaker, and the control is performed so as to lower the oscillation frequency of the inverter circuit 2.

Therefore, in this device, the impedance control amount when one lamp is turned on and one lamp is not turned on is Z4, which is the same as when both lamps are not turned on. From the oscillation frequency characteristic g4, the oscillation frequency becomes f (AND). At this time, the tube voltage VL applied to the unlit fluorescent lamp is the frequency-tube voltage characteristic g1 in FIG.
To VL11, which is considerably higher than the originally required tube voltage VL1.

[0019]

As described above, in the conventional device shown in FIG. 6, when the two lamps are lit or the two lamps are not lit, the proper oscillation frequency can be controlled by the proper impedance control, and each fluorescent lamp can be controlled. The tube voltage applied to the lamp can be controlled appropriately, but when one lamp is lit and one lamp is not lit, an appropriate tube voltage cannot be applied to the fluorescent lamp on the non-lit side, and the fluorescent lamp on the non-lit side is quickly lit. There was a problem I could not do. Further, in the conventional device shown in FIG. 9, when two lights are turned on or two lights are not turned on, an appropriate oscillation frequency can be controlled by an appropriate impedance control, and the tube voltage applied to each fluorescent lamp can be appropriately controlled. When one lamp is turned on and one lamp is not turned on, an excessive tube voltage is applied to the fluorescent lamp on the non-lit side, which causes a problem of shortening the life of the fluorescent lamp on the non-lit side.

Therefore, according to the invention of claim 1, not only when all the discharge lamps are in the non-lighting state and when all the discharge lamps are in the lighting state, but when only some of the discharge lamps are in the lighting state. Also in the above, there is provided a discharge lamp lighting device capable of controlling an appropriate oscillation frequency by appropriate impedance control and capable of always properly controlling a tube voltage applied to a discharge lamp.

[0021]

According to the first aspect of the present invention,
In a discharge lamp lighting device for lighting a plurality of discharge lamps in parallel using an inverter circuit that performs self-excited oscillation control, the oscillation frequency of the inverter circuit is variably controlled by the impedance control amount of a feedback circuit for self-excited oscillation control. Then, the impedance control amount changes to the first control amount when all the discharge lamps are lit, and to the second control amount when all the discharge lamps are in the non-lighting state, and is different from the lit discharge lamps. When there are mixed discharge lamps, the third
The control frequency of the inverter circuit is controlled so that an appropriate voltage for maintaining lighting is applied to each discharge lamp by the first control amount, and each emission voltage is controlled by the second control amount. The oscillation frequency of the inverter circuit is controlled so that an appropriate voltage is applied to the electric lamp to start lighting, and a voltage appropriate to start lighting the non-lighted discharge lamp by the third control amount. Is applied to control the oscillation frequency of the inverter circuit.

[0022]

Embodiments of the present invention will be described below with reference to the drawings. As shown in FIG.
The inverter circuit 52 is connected to. DC power supply 5
Reference numeral 1 is composed of, for example, a circuit that performs full-wave rectification and smoothing of an AC power supply, and the negative electrode side is grounded.

The inverter circuit 52 is a DC power source 51.
A pair of MOS type FET (field effect transistor) 5
A series circuit of 3, 54 is connected in parallel, a series circuit of a resistor 55 and a capacitor 57 is connected between the gate and the source of the FET 53, and a resistor 5 is connected between the gate and the source of the FET 54.
The series circuit of 6 and the capacitor 58 is connected. The connection point of the FETs 53 and 54 is connected to one end of the first current transformer 59, and the series circuit of the resistor 55 and the capacitor 57 is connected via the resistor 60 to the first circuit.
The second current transformer 61 magnetically coupled to the current transformer 59 of No. 3 is connected in parallel, and the third current magnetically coupled to the first current transformer 59 is connected to the series circuit of the resistor 56 and the capacitor 58 via the resistor 62. The current transformer 63 is connected in parallel.

The other end of the first current transformer 59 is connected to one end of one filament electrode 66a of the first fluorescent lamp 66 through the capacitor 64 and the first choke coil 65 in series, and the capacitor 67 is also connected. And the second choke coil 68 are connected in series to one end of one filament electrode 69a of the second fluorescent lamp 69. One end of the other filament electrode 66b, 69b of each fluorescent lamp 66, 69 is connected to the source of the FET 54. A capacitor 70 is connected between the other ends of the filament electrodes 66a and 66b of the fluorescent lamp 66, and a capacitor 71 is connected between the other ends of the filament electrodes 69a and 69b of the fluorescent lamp 69.

A series voltage dividing circuit of capacitors 72 and 73 is connected between one ends of the filament electrodes 66a and 66b of the fluorescent lamp 66, and a capacitor 7 is connected between one ends of the filament electrodes 69a and 69b of the fluorescent lamp 69.
4,75 connected in series voltage divider circuit,
The voltage dividing points of 2, 73 are NP through the constant voltage diode 76.
It is connected to the base of the N-type transistor 77, and the voltage dividing point of the capacitors 74 and 75 is a constant voltage diode 78.
It is connected to the base of the NPN type transistor 79 via.

Also, a resistor 80,
The PNP transistor 81 has an emitter and a collector connected in series with a resistor 82 and a resistor 83 and a resistor 8
A series circuit of 4, 85 and a resistor 86 is connected. The base of the transistor 81 is connected to the connection point between the resistors 84 and 85. The collector and the emitter of the transistor 77 are connected in parallel to the resistor 86 via the resistor 87, and the collector and the emitter of the transistor 79 are connected in parallel via the resistor 88. As shown in FIG. 2, the + VE voltage is a resistance 89 between the input terminal I of the full-wave rectified output obtained by full-wave rectifying the AC power supply and the ground.
It is obtained from a power supply circuit in which a smoothing capacitor 90 is connected via a constant voltage diode 91 having a Zener voltage of + VE connected in parallel to the smoothing capacitor 90.

Further, a fourth current transformer 92 magnetically coupled to the first current transformer 59 is provided, and one end of the fourth current transformer 92 is connected to the cathode of the diode 94 and the anode of the diode 95 via the capacitor 93. The other end of the fourth current transformer 92 is connected to the anode of the diode 94 and grounded. The diode 9 to the diode 94
Capacitor 96 is connected in parallel via 5. The drain and source of a MOS FET (field effect transistor) 97 are connected in parallel to the capacitor 96,
The gate of the FET 97 is connected to the connection point of the resistors 82 and 83.

In such a configuration, the capacitor 6
4, each FET 53, 54 at a frequency close to the resonance frequency with respect to the resonance system of the series resonance circuit of the first choke coil 65 and the capacitor 70 and the series resonance circuit of the capacitor 67, the second choke coil 68 and the capacitor 71. Resonance is sustained by switching alternately. When the inverter circuit 52 operates, an alternating current flows through the first current transformer 59, and a part of this current also flows through the magnetically coupled second to fourth current transformers 61, 63, 92. The second and third current transformers 61 and 63 have opposite phases, and the self-excited oscillation is maintained by feeding back the resonance current phase of itself.

Further, each current transformer 59, 61, 6
Since 3 and 92 utilize supersaturated magnetism, the oscillation frequency of the inverter circuit 52 can be controlled by adjusting the saturation amount. That is, the voltage generated in the fourth current transformer 92 is increased through the voltage doubler rectifying circuit of the capacitor 93 and the diodes 94 and 95 to charge the capacitor 96, but the charge of the capacitor 96 is extracted by the FET 97. At this time, the impedance of the fourth current transformer (92) system is adjusted by controlling the impedance of the FET 97 to control the amount of electric charge extracted from the capacitor 96.

For example, the fourth current transformer (92)
When the impedance of the system is lowered, the saturation degree of the first current transformer 59 magnetically coupled to this is lowered,
The off time of the gate voltage applied to the second and third current transformers 61 and 63 is delayed. As a result, the self-excited oscillation frequency of the inverter circuit 52 is lowered, and the resonance point is approached in both the lighting mode and the non-lighting mode, so that the resonance is increased. That is, in the lighting mode, the lamp current increases and becomes brighter, and in the non-lighting mode, the tube voltage rises and the lighting becomes easier.

On the contrary, if the impedance of the fourth current transformer (92) system is increased, the saturation degree of the first current transformer 59 magnetically coupled to the fourth current transformer (92) increases, and the second,
The off time of the gate voltage applied to the third current transformers 61 and 63 is shortened. As a result, the inverter circuit 52
Since the self-excited oscillating frequency rises and moves away from the resonance point in both the lighting mode and the non-lighting mode, the resonance becomes small.

The current flowing through the FET 97 is the resistance 80, 8
It can be controlled by determining the gate potential of the FET 97 with a circuit composed of 2, 83, 84, 85, 86 and the PNP type transistor 81. When the power is turned on, the potential of the smoothing capacitor 90 of the power supply circuit that generates + VE voltage is zero, rises with the passage of time, and stabilizes when the zener voltage of the constant voltage diode 91 is reached. In the process of increasing the voltage, the gate-source voltage VGS of the FET 97 is also shown in FIG.
As shown in (a), the voltage gradually rises and remains in the off state until the threshold level of the FET 97 is exceeded, and when it exceeds the threshold level, the FET 97 gradually shifts to the on state. When the FET 97 shifts to the ON state, the control impedance Z changes as shown in FIG. 3 (b). That is, the control impedance is gradually lowered, and thereby the oscillation frequency of the inverter circuit 52 is lowered, so that the tube voltage applied to each of the fluorescent lamps 66 and 69 is higher than the initial state.

It is preset so that the constant voltage diodes 76, 78 are rendered conductive by the voltage dividing point of the capacitor with respect to the increased tube voltage. That is, the tube voltage applied to the fluorescent lamp 66 is divided by the capacitors 72 and 73,
When this voltage dividing point voltage exceeds the Zener voltage of the constant voltage diode 76, the constant voltage diode 76 becomes conductive. In addition, the tube voltage applied to the fluorescent lamp 69 is the capacitor 7
The voltage is divided by 4, 75, but the voltage dividing point voltage exceeds the Zener voltage of the constant voltage diode 78, so that the constant voltage diode 78 becomes conductive. When the constant voltage diode 76 is turned on, the transistor 77 is turned on, and when the constant voltage diode 78 is turned on, the transistor 79 is turned on.

Thus, the resistances 87 and 8 are different from the resistance 86.
8 are connected in parallel, and the combined resistance value of this parallel circuit decreases with respect to the resistance value of the resistor 86 alone. Therefore, the base potential of the transistor 81 decreases, and the transistor 8
The base current of 1 increases. As a result, the impedance between the emitter and collector of the transistor 81 decreases, and the gate potential of the FET 97 increases. Thus, F
The control impedance between the drain and the source of the ET 97 is lowered, the charge of the capacitor 96 is extracted, and the impedance of the fourth current transformer 92 is lowered. As a result, the magnetic saturation of the first current transformer 59 is weakened, and as a result, control is performed so as to reduce the oscillation frequency of the inverter circuit 52.

Thus, the fluorescent lamp 6 at the time of starting
In the non-lighted state of Nos. 6 and 69, the impedance control amount is determined so that a predetermined tube voltage VL appropriate for starting the lighting of the fluorescent lamp can be applied. That is,
The combined resistance value of the resistors 86, 87, 88 is determined.

When both of the fluorescent lamps 66 and 69 are turned on, the tube voltage VL is lowered, so that the constant voltage diodes 76 and 78 become non-conductive and the transistors 77 and 7 are turned on.
9 also turns off. In this way, the resistors 87 and 88 are electrically separated from the resistor 86. At this time, the fluorescent lamp 6
Appropriate tube voltage VL is applied to 6 and 69, and the resistance value of the resistor 86 is determined so that the lighting illuminance of the fluorescent lamps 66 and 69 becomes a predetermined brightness.

Further, for example, when one lamp is turned on and one lamp is not turned on, for example, when the fluorescent lamp 66 is turned on and the fluorescent lamp 69 is not turned on, the constant voltage diode 76 becomes non-conductive. The constant voltage diode 78 maintains the conductive state. Therefore, in this case, only the transistor 79 is turned on, and only the resistor 88 is connected in parallel to the resistor 86. Therefore, the base potential of the transistor 81 is determined by this combined resistance value, and the base current corresponding thereto flows, which determines the gate potential of the FET 97. At this time, the combined resistance value of the resistors 86 and 88 is determined so that an appropriate tube voltage VL is applied to the fluorescent lamp 69 on the non-lighting side.
On the contrary, when the fluorescent lamp 69 is turned on and the fluorescent lamp 66 is not turned on, the same operation is required. Therefore, the resistance values of the resistors 87 and 88 are made equal.

The graph of FIG. 4 shows the frequency of the resonance mode during non-lighting-the tube voltage characteristic g1 and the frequency of the resonance mode during lighting-
The tube voltage characteristic g2 is shown. It should be noted that f0 is the resonance frequency when the fluorescent lamp is lit, and f1 is the resonance frequency when the fluorescent lamp is not lit. In the graph of FIG. 5, the impedance control amount-oscillation frequency characteristic g3 when both the fluorescent lamps 66 and 69 are not lit, the impedance control amount when one lamp is lit and the one lamp is not lit-oscillation frequency characteristic g4, 2 The figure shows the impedance control amount-oscillation frequency characteristic g5 when both lights are on. Each graph shows the change of the impedance control amount and the change of the oscillation frequency when the control impedance Z changes as Z1 → Z4 → Z3 → Z2 in the starting sequence as shown in FIG. 3 (b).

In this starting sequence, first, the control impedance Z starts to operate in the non-lighting mode with the controlled amount of Z1 and the preheating frequency at that time is f (preheating). When the frequency is f (preheat), the fluorescent lamp 66,
The tube voltage applied to 69 is the frequency-tube voltage characteristics g1 to VL2 of the resonance mode during non-lighting. This voltage is a voltage at which the lamp does not discharge, and LC resonance of the capacitor 64, the choke coil 65 and the capacitor 70 is caused by the fluorescent lamp 66.
An electric current is passed through the filament electrodes 66a and 66b to preheat the filament electrodes 66a and 66b, and the LC resonance of the condenser 67, the choke coil 68, and the condenser 71 passes through the filament electrodes 69a and 69b of the fluorescent lamp 69. Then, an electric current is supplied to preheat the filament electrodes 69a and 69b. Thus, first, the filament electrodes of the fluorescent lamps 66 and 69 are preheated.

Next, when the control impedance Z is the lowest and becomes Z4, the impedance control amount when the two lights are not lit-
From the oscillation frequency characteristic g3, the oscillation frequency of the inverter circuit 52 decreases to f (pressurization). The tube voltage applied to the fluorescent lamps 66 and 69 at this time is VL1 from the frequency-tube voltage characteristic g1 of the resonance mode during non-lighting, and this voltage is an appropriate voltage for starting and lighting the lamp. There is.

In this way, since a proper tube voltage is applied to the fluorescent lamps 66 and 69 to start and light them, discharge of both lamps starts, but here, the fluorescent lamp 66 is temporarily turned on first, and the fluorescent lamp 69 is lit. When the fluorescent lamp 66 is turned on with a delay, the control amount of the control impedance Z changes from Z4 to Z3 when the fluorescent lamp 66 is turned on. That is, the control impedance Z increases. The oscillation frequency of the inverter circuit 52 at this time is from the impedance control amount-oscillation frequency characteristic g4 to f when one lamp is lit and one lamp is not lit.
(Pressurization). That is, the two lamps have the same frequency as the oscillation frequency when they are not lit. Therefore, the tube voltage applied to the non-lighted fluorescent lamp 69 also becomes VL1 from the frequency-tube voltage characteristic g1 of the resonance mode in the non-lighted state, and becomes the same voltage as when the two lamps are not lit.

As described above, the tube voltage applied to the fluorescent lamp 69 does not change even if the fluorescent lamp 66 is turned on first, so that the fluorescent lamp 69 can also be quickly started and turned on. When the fluorescent lamp 69 is also turned on, the control amount of the control impedance Z changes from Z3 to Z2. The oscillation frequency of the inverter circuit 52 at this time is from the impedance control amount-oscillation frequency characteristic g5 to f when two lights are lit.
(Lights up). The tube voltage applied to the fluorescent lamps 66, 69 changes from the frequency-tube voltage characteristic g2 of the resonance mode during lighting to VL3. That is, the fluorescent lamps 66, 6
The voltage is appropriate to maintain the lighting of No. 9.

As described above, when the two lights are not lit, the control amount of the control impedance Z is the second control amount Z4.
At this time, the oscillation frequency of the inverter circuit 52 is f
(Pressurization), and a proper tube voltage VL1 is applied to the fluorescent lamps 66 and 69 to start lighting. And 1
When one lamp is lit and one lamp is not lit, the controlled variable of the control impedance Z becomes the third controlled variable Z3, and at this time, the oscillation frequency of the inverter circuit 52 becomes f (pressurized) and the unlit fluorescent light is emitted. The tube voltage VL1 appropriate to start lighting is continuously applied to the lamp. Further, when both lamps are turned on, the control amount of the control impedance Z becomes the first control amount Z2, at which time the oscillation frequency of the inverter circuit 52 becomes f (lighting), and the fluorescent lamps 66, 69 are turned on. Appropriate tube voltage VL3 is applied to maintain.

Therefore, of course, when the two fluorescent lamps 66 and 69 are not lit and when the two fluorescent lamps 66 and 69 are lit, one is lit and one is unlit. Also in this case, the appropriate oscillation frequency can be controlled by appropriate impedance control, and the tube voltage constantly applied to the fluorescent lamp can be appropriately controlled.

By the way, in the case where one is lit and one is not lit, it is desirable to raise the tube voltage slightly higher than the initial tube voltage VL1 in order to surely bring the unlit fluorescent lamp to the lighting. . To realize this, the control amount of the control impedance Z may be slightly shifted to the Z4 side from Z3.
That is, when the controlled variable deviates from Z3 to Z4, the oscillation frequency of the inverter circuit 52 shifts from f (pressurization) to f (lighting), so that the tube voltage applied to the non-lighting fluorescent lamp is higher than VL1. Will also be slightly raised, and the unlit fluorescent lamp can be reliably turned on.

To slightly shift the control amount of the control impedance Z to the Z4 side from Z3, the resistance values of the resistors 86, 87 and 88 are adjusted. That is, the resistance value of the resistor 86 is determined so as to obtain the tube voltage VL3 that provides a predetermined illuminance when the two lamps are lit, and when the two lamps are not lit, the tube voltage VL1 is appropriate to start the lighting and one lamp is lit. The resistance 86, so that the tube voltage is slightly higher than the tube voltage VL1 when one lamp is not lit,
The resistance values of 87 and 88 may be determined. In the above-described embodiment, the case where two fluorescent lamps are lit in parallel has been described. However, the present invention is not limited to this, and when three, four, or five or more fluorescent lamps are lit in parallel. Can also be applied. Further, in the above-described embodiment, the one using the fluorescent lamp as the discharge lamp has been described, but the present invention is not necessarily limited to this.

[0047]

As described above, according to the invention described in claim 1, not only when all the discharge lamps are in the non-lighting state and when all the discharge lamps are in the lighting state, but only some of the discharge lamps are turned on. Even in the state, the appropriate oscillation frequency can be controlled by the appropriate impedance control, and the tube voltage applied to the discharge lamp can always be properly controlled. That is, when a plurality of discharge lamps are lit in parallel, the discharge lamp that is easy to light is first lit, and the discharge lamp that is hard to light remains as non-lighting, but in such a case, it is applied to the non-lighting discharge lamp. Since the tube voltage continues to be maintained properly, the non-lighted discharge lamp can be reliably turned on. Further, since the excessive tube voltage is not applied to the non-lit discharge lamp, it is possible to prevent the life of the discharge lamp from being shortened.

[Brief description of drawings]

FIG. 1 is a circuit configuration diagram showing an embodiment of the present invention.

FIG. 2 is a power supply circuit diagram for generating a + VE voltage in the same embodiment.

FIG. 3 is a diagram showing a rising characteristic of a gate-source voltage and a changing characteristic of a control amount of a control impedance of the FET when the power is turned on in the embodiment.

FIG. 4 is a diagram showing a frequency-tube voltage characteristic in a non-lighting resonance mode and a frequency-tube voltage characteristic in a lightning resonance mode in the same embodiment;

FIG. 5 is a diagram illustrating an impedance control amount when two lamps are not lit-oscillation frequency characteristic, an impedance control amount when one lamp is lit, and one lamp is not lit-oscillation frequency characteristic and an impedance control amount when two lamps are lit; The figure which shows an oscillation frequency characteristic.

FIG. 6 is a circuit configuration diagram showing a conventional example.

FIG. 7 is a diagram showing a frequency-tube voltage characteristic of a non-lighting resonance mode and a frequency-tube voltage characteristic of a lightning resonance mode in the conventional example.

FIG. 8: Impedance control amount when two lamps are not lit-oscillation frequency characteristic, impedance control amount when one lamp is lit, one lamp is not lit-oscillation frequency characteristic and impedance control amount when two lamps are lit-oscillation in the conventional example The figure which shows a frequency characteristic.

FIG. 9 is a circuit configuration diagram showing another conventional example.

FIG. 10 is a view showing a frequency-tube voltage characteristic of a non-lighting resonance mode and a frequency-tube voltage characteristic of a lightning resonance mode in the conventional example.

FIG. 11: Impedance control amount when two lamps are not lit-oscillation frequency characteristic, impedance control amount when one lamp is lit, one lamp is not lit-oscillation frequency characteristic and impedance control amount when two lamps are lit-oscillation in the conventional example The figure which shows a frequency characteristic.

[Explanation of symbols]

52 ... Inverter circuit 59, 61, 63, 92 ... Current transformer 64, 67, 70, 71 ... Capacitor 65, 68 ... Choke coil 66, 69 ... Fluorescent lamp 72, 73, 74, 75 ... Capacitor 76, 78 ... Constant voltage Diode 77, 79, 81 ... Transistor 80, 82, 83, 84, 85, 86, 87, 88 ... Resistor 96 ... Capacitor 97 ... MOS type FET (field effect transistor)

Claims (1)

[Claims] 1. A discharge lamp lighting device for lighting a plurality of discharge lamps in parallel using an inverter circuit for performing self-excited oscillation control, wherein impedance control of a feedback circuit for self-excited oscillation control of an oscillation frequency of the inverter circuit. The impedance control amount changes to a first control amount when all the discharge lamps are lit, and a second control amount when all the discharge lamps are in a non-lighting state. , When a lit discharge lamp and a non-lit discharge lamp are mixed, the third controlled variable is applied, and a proper voltage is applied to each of the discharged lamps by the first controlled variable to maintain lighting. The oscillation frequency of the inverter circuit is controlled so that the voltage of the inverter circuit is controlled so as to apply a proper voltage to the lighting of each discharge lamp by the second control amount. The discharge lamp is characterized in that the oscillation frequency is controlled, and the oscillation frequency of the inverter circuit is controlled so that an appropriate voltage is applied to the non-lighted discharge lamp to start lighting by the third control amount. Lighting device.