JPH088156B2 - Discharge lamp lighting device - Google Patents

Description

TECHNICAL FIELD The present invention relates to a discharge lamp lighting device using a semiconductor switching element.

(Background Art) In a conventional general discharge lamp lighting device, since the ballast element is configured by a choke coil, a transformer, a capacitor, or the like, or by a combination thereof, both dimensions and weight are improved. It was great. From this point, there is a demand for downsizing, weight reduction, and high efficiency of the discharge lamp lighting device, and therefore, it has been proposed to light the discharge lamp at a high frequency. For example, as a lighting device for a fluorescent lamp, a high-frequency lighting device using a switching transistor, a thyristor or the like has already been put into practical use.

On the other hand, if the high-pressure discharge lamp lighting device is also subjected to high-frequency lighting, as in the case of the fluorescent lamp lighting device, it is possible to obtain the effects of downsizing and weight reduction, high efficiency, etc. It is conventionally known that there is arc instability due to an acoustic resonance phenomenon. Here, the formation mechanism of the instability of the arc that occurs when the high-pressure discharge lamp is lit at high frequencies includes high-frequency fluctuations in the electrical input, pressure changes in the gas inside the arc tube, the generation of standing pressure waves at special frequencies, and pressures above the limit. It is considered that the arc is unstable due to the amplitude.

The "special frequency" is a so-called acoustic resonance frequency, which is determined by the dimension of the arc (actually, the shape of the arc tube) and the speed of sound in the arc tube, and the sound speed is the average of the gas. Since it is determined by the molecular weight and the ion temperature, if these values are known, it can be determined relatively easily. Further, since the acoustic resonance frequency at which the "arc instability due to the pressure amplitude exceeding the limit" occurs is a problem in the non-linear region, the answer cannot be simply obtained.

By the way, as a method of eliminating the instability of the arc caused by the acoustic resonance phenomenon, the oscillation frequency of the inverter circuit that becomes the lighting power source of the discharge lamp is set to an ultrahigh frequency (for example, 100 KHz) (Japanese Patent Laid-Open No. 56-11897). However, in this conventional example, noise (especially radiation noise) is large and switching loss is large, so that there is a problem in practical use. There is also a method of modulating the frequency of the lighting power supply between 30 KHz and 50 KHz (for example, Japanese Patent Laid-Open No. 56-48095), but the instability of the arc cannot be suppressed and there is still a practical problem. . Further, there is a lamp which is lit by direct current (Japanese Patent Laid-Open No. 56-98499), but this also has a problem in practical use in terms of lamp life (blackening, etc.). In addition, it is based on the lighting of a square wave. (For this square wave lighting, refer to "Characteristics of Acoustic Resonance in Discharge Lamps (Characte
risticis of Acoustical Resonance In Discharge Lamp
s) ”Illuminating Engineering (ILLUMIN
ATING ENGINEERING) December 1970, P713-716. ) Is also available. However, with this method, the arc is stable, but in the flat part of the rectangular wave, it becomes large because it is burdened by the resistance as a current limiting element, and the power loss becomes large. Since the rise and fall are suddenly completed, there arises a problem that noise is generated, and there is a problem that this measure leads to a significant increase in cost.

Therefore, basically, it is most practical to operate the current limiting element portion under a high frequency while making the lamp current flowing through the high pressure discharge lamp a low frequency to prevent an acoustic resonance phenomenon. As a conventional example adopting such a method, there is an apparatus disclosed in Japanese Utility Model Laid-Open No. 59-16100. In this device, a high frequency component is superposed on a flat portion of a rectangular wave, and an inductance is used as a current limiting element to miniaturize the device. However, this method requires a transformer for oscillation and a semiconductor element dedicated to the chopper, and the lighting device is still large.

There is also a device disclosed in US Pat. No. 4,170,747.
FIG. 5 is a circuit diagram of a conventional discharge lamp lighting device disclosed in US Pat. No. 4,170,747. In the figure capacitor
The power supply terminals 1 and 2 connected to Ca connected via a transistor Tr ₁ to Tr ₄ a switching circuit the current detection resistor Rs connected to the bridge type, the transistor Tr ₁ to Tr ₄ each transistor Tr ₁ ~ Diode D _{1 in the} opposite direction of Tr ₄
~ D ₄ are connected in parallel, and the connection point of transistors Tr ₁ and Tr ₄
A capacitor C should be placed between the connection point of the transistors Tr ₂ and Tr _3.
A series circuit of a high pressure discharge lamp DL in which b is connected in parallel and a choke coil L is connected. The control circuit A uses the transistors Tr ₁ to Tr ₄ according to the current detected by the current detection resistor Rs.
Is a circuit for controlling.

FIGS. 6A to 6D are timing charts showing the operations of the transistors Tr _{1 to} Tr ₄ . Transistors Tr _1, Tr ₂ are turned on and off at a low frequency of about 50 / 60Hz or 400Hz by the control circuit A, when the transistor Tr ₁ is turned on performs on-off operation of the transistor Tr ₃ is e.g. 20KHz duty of the variable, the transistor Tr ₂ When is on, the transistor Tr ₄ performs an on / off operation similar to the transistor Tr ₃ . Further, when the polarity of the high pressure discharge lamp DL is reversed, the four transistors Tr _{1 to} Tr ₄ are simultaneously turned off during the idle period T _D
Is provided. The control circuit A changes the on / off duty ratio depending on the magnitude of the current detection value by the current detection resistor Rs.

The operation of the circuit of FIG. 5 will be further described with reference to FIGS. 6 (a) to 6 (d). In the period from t ₁ to t _{2 in} FIG. 6, a transistor Tr operating at a low frequency as shown in FIG. 6 (a). ₁ is an ON operation state, and the transistor Tr ₃ is in an ON / OFF operation state at a high frequency as shown in FIG. And this t ₁
In between ~t _2, transistor Tr _2, Tr ₄ as shown in FIG. (B) (d) is in the off state. When the transistor Tr ₃ is turned on, a power supply, a power supply terminal 1, a transistor Tr ₁ , a choke coil L, a high pressure discharge lamp DL and a parallel circuit of a capacitor Cb,
Transistor Tr ₃ , current detection resistor Rs, power supply terminal 2, closed circuit of power supply is formed, current flowing in choke coil L
I _L rises linearly with a constant slope,
When Tr ₃ turns off, the direction in which the stored energy determined by the current I _L at this time and the inductance value of the choke coil L tries to keep the current flowing, that is, the transistor
Current flows in the same direction as when the Tr ₃ is on,
The stored energy is released by the closed circuit of the choke coil L, the high pressure discharge lamp DL and the capacitor Cb, the diode D ₂ , the transistor Tr ₁ and the choke coil L, and the above operation is repeated until time t ₂ .

In a period T _D of the following t ₂ ~t _3, 4 pieces of transistors
Since all of Tr _{1 to} Tr ₄ are off, no power is supplied from the power supply connected to the power supply terminals 1 and 2 during this idle period.

Further, during the period of t _{3 to} t ₄ , basically the same operation as the period of t _{1 to} t ₂ is performed, but it is the transistors Tr ₂ and Tr ₄ that operate during this period, and the transistors Tr ₁ , Tr ₃ is off. When the transistor Tr ₄ turns on, the power supply, the power supply terminal 1, the transistor Tr ₂ , the parallel circuit of the high-pressure discharge lamp DL and the capacitor Cb, the choke coil L, the transistor Tr ₄ , the current detection resistor Rs, the power supply terminal 2, the power supply closed circuit. When the transistor Tr ₄ is turned off, a current I _L flows through the choke coil L, the diode D ₁ , the transistor Tr ₂ , the parallel circuit of the high-pressure discharge lamp DL and the capacitor Cb, and the closed circuit of the choke coil L. Here, the current I _L shown in FIG.
Is opposite to the direction in which the current I _L, the orientation of Ila is illustrated in the direction of the current Ila to flow to the high-pressure discharge lamp DL is shown for the period t ₁ ~t _2, the period of t ₃ ~t ₄ and . The diode
D ₃ and D ₄ do not flow current in normal operation, but are provided to flow a surge current during a transition.

In the next period of t ₄ to t ₅ , 4 transistors Tr
_{Since 1 to} Tr ₄ are all off, power is not supplied from the power supply connected to the power supply terminals 1 and 2 during this idle period. The quiescent period T _D (t _{2 to} t ₃ , t _{4 to} t ₅ ) is the transistor T
r _1, Tr ₄ or Tr _2, Tr ₃ presents a short-circuit state simultaneously turned on, the lighting device is a dead time for preventing from reaching the breakdown. This simultaneous turn-on occurs when the storage time becomes long due to variations in transistor elements and temperature rise, and due to timing shift between transistors.

As described above, a rectangular wave-shaped alternating current containing high frequency ripples flows through the high pressure discharge lamp DL.

FIG. 7 is a timing diagram showing an enlarged time scale in the vicinity of t ₂ to t _{3 in} FIG. 6, and FIGS. 7A and 7B show the on / off states of the transistors Tr ₃ and Tr ₄ , respectively. The waveforms of the current I _L , the current Ila, and the voltage Vla across the high-pressure discharge lamp DL are shown in FIGS. (C), (d), and (e). When the transistor Tr ₃ is turned off as shown in the diagram (a) at time t _2, the charge of the capacitor Cb is released the high-pressure discharge lamp DL, current Ila flows with a slight delay compared with the current I _L But,
Finally the current Ila becomes zero. In other words, the current Ila, which is the lamp current, is paused. Next, at time t ₃ , the polarity of the current Ila changes, and the transistor Tr ₄ operates as shown in FIG. Since current Ila is zero immediately before the t ₃ time, the voltage across Vla of the high-pressure discharge lamp DL increases just after the time point of t _3. This is a phenomenon peculiar to the discharge lamp, and the current Il
When a becomes zero, the ions in the arc tube disappear, and at time t ₃ , a high voltage is required across the high-pressure discharge lamp DL to maintain lighting. This voltage is the so-called restrike voltage. If the period during which the current Ila is zero becomes too long, lighting may not be maintained even if a voltage is applied at the time point t ₃ , and the lamp may go out. Further, the re-ignition voltage becomes high even if it does not go out, and flicker may occur when it reaches near the power supply voltage. In the case of a fluorescent lamp, the above phenomenon is alleviated a little, but in the case of a high-pressure discharge lamp, the re-ignition voltage rises significantly even if some rest occurs.

As described above, the quiescent period T _D is provided to prevent the transistors Tr _{1 to} Tr ₄ from turning on at the same time, but this causes the current Ila to quiesce and the re-ignition voltage to rise, causing flicker. In the worst case, it could disappear. Further, the rise of the re-ignition voltage may accelerate the consumption of the electrodes and shorten the life of the lamp, and there is also a problem that the lamp may be extinguished due to a sudden decrease in the power supply voltage. In particular, the higher the wattage discharge lamp, the lower the wattage, the greater the degree of increase in the re-ignition voltage due to the suspension of the current Ila, which is troublesome in the circuit design.

Therefore, the transistors Tr _{1 to} Tr ₄ are turned on at the same time during the period T _D.
As a method of solving the above problems while preventing with,
There is a method of increasing the capacity of the capacitor Cb connected in parallel to the high pressure discharge lamp DL so that the current Ila does not become zero in the period T _D. In this case we consider the high-pressure discharge lamp DL equivalent resistance and a time constant of the CR, in relation to the period T _D, the capacitor
The capacitance of Cb can be selected to an appropriate value, but in this case, the capacitance becomes several tens of times, the capacitor Cb becomes large, and there is a problem that the cost is significantly increased.

In order to solve the above problems, the present inventors have devised a circuit as shown in FIG. The circuit of FIG. 8 has a so-called full bridge structure, and the transistors Tr ₅ , Tr ₆ , Tr ₇ , Tr ₈ and the diodes D ₅ , D ₆ , D ₇ , D ₈ used as switching elements are the circuit of FIG. Transistor of Tr ₃ , Tr ₄ , T
The basic operation is the same as that of r ₁ , Tr ₂ and diodes D ₃ , D ₄ , D ₁ , D ₂ . In the circuit of FIG. 5, the transistors Tr ₃ and Tr ₄ that switch at high frequencies are on the negative side of the power supply, and the transistors Tr ₁ and Tr ₂ that switch at low frequencies are on the positive side of the power supply.
On the other hand, in the circuit of FIG. 8, transistors Tr ₅ and Tr ₆ that switch at high frequency are provided on the + side of the power supply, and transistors Tr ₇ and Tr ₈ that switch at low frequency are provided on the − side of the power supply. The difference lies in that the capacitor Cb is connected in parallel to the series circuit of the high pressure discharge lamp DL and the inductance element L ₂ . DC power supply is AC power supply AC full-wave rectified by full-wave rectifier DB and smoothing capacitor Ca
It is obtained by smoothing with and supplied to the circuit.

Thus, in the circuit of FIG. 8, when the transistor Tr ₇ is on and the transistors Tr ₆ and Tr ₈ are off, when the transistor Tr ₅ is turned on, the power supply, the power supply terminal 1, the transistor Tr ₅ and the choke coil are turned on. L, a circuit in which a capacitor Cb is connected in parallel to a series circuit of an inductance element L ₂ and a high-pressure discharge lamp DL, a transistor Tr ₇ , a current detection element Zs, a power supply terminal 2, a current flows through a power supply path, and a transistor Tr ₅ When is turned off, a current flows in the path of the circuit in which the capacitor Cb is connected in parallel to the series circuit of the choke coil L, the inductance element L ₂ and the high pressure discharge lamp DL, the transistor Tr ₇ , the diode D ₈ and the choke coil L.
When transistors Tr ₅ and Tr ₇ are off and transistor Tr ₈ is on, transistor Tr ₆
When is turned on, power supply, power supply terminal 1, transistor Tr ₆ ,
A circuit in which a capacitor Cb is connected in parallel to a series circuit of a high pressure discharge lamp DL and an inductance element L ₂ , a choke coil L, a transistor Tr ₈ , a current detection element Zs, a power supply terminal 2,
When current flows through the power supply path and transistor Tr ₆ turns off, choke coil L, transistor Tr ₈ , diode D
_{7. A} current flows through the path of the choke coil L, the circuit in which the capacitor Cb is connected in parallel to the series circuit of the high pressure discharge lamp DL and the inductance element L ₂ .

The capacitor Cb has the polarity shown in FIG. 8 at the time when the transistor Tr ₅ ends the on-state (at the time corresponding to t ₂ in FIG. 6), and in the rest period T _D thereafter, the capacitor Cb and the inductance element. In the closed circuit of L ₂ and the high-pressure discharge lamp DL, the charge of the capacitor Cb is released to generate an oscillating current, and the current Ila flowing in the high-pressure discharge lamp DL is inverted while maintaining continuity, and the transistor Tr ₆ is turned on. At the time when the state is started (at the time corresponding to t ₃ in FIG. 6),
The inversion is completed without causing a pause in the current Ila. The voltage Vla across the high-pressure discharge lamp DL also has a waveform similar to the current Ila, and the re-ignition voltage becomes minimal. Thus, by connecting the inductance element L ₂ having a small inductance value in series with the high pressure discharge lamp DL, the high pressure discharge lamp DL can be maintained in the excited state during the quiescent period T _D , and the conventional problems are solved. be able to.

In addition, the inductance element L ₂ is ₂ of the pulse transformer PT.
The secondary winding L ₂ is used. Then, in the series circuit of this inductance element L ₂ and high pressure discharge lamp DL, a capacitor
Connect a series circuit of C ₁ and resistor R ₁ and connect a series circuit of primary winding L ₁ of pulse transformer PT and bidirectional three-terminal thyristor Q _{1 in} parallel with capacitor C ₁ to form an igniter. The circuit G is configured. In this igniter circuit G, the bidirectional three-terminal thyristor Q ₁ receives a signal from the trigger circuit A ₄ in the control circuit A, is periodically turned on, and the charge of the capacitor C ₁ charged via the resistor R ₁ is Capacitor
C _1, a pulse transformer PT of the primary winding L _1, by releasing suddenly竣the circuit of bi-directional triode thyristor Q _1, determined by the secondary winding L ₂ the winding ratio of the pulse transformer PT A high-voltage pulse is induced and applied to the high-pressure discharge lamp DL via the capacitor Cb to start it.

The control circuit A is a high-frequency oscillator circuit A ₁ , a low-frequency oscillator circuit A ₂ ,
Flip-flop circuit A _3, the trigger circuit A _4, and a base drive circuit A ₅ to A ₈ and NOR gate A _9, A ₁₀ of the transistor Tr ₅ to Tr _8. FIG. 9 shows a specific example of the low-frequency oscillator circuit A ₂ and the flip-flop circuit A _3. The low-frequency oscillator circuit A ₂ is a timer circuit tm composed of a general-purpose timer IC (eg, μPC1555 manufactured by NEC Corporation).
And external resistors R ₂ and R ₃ for setting the time constant and capacitor C
_{It consists of 2} and the cycle determined by the set time constant.
A pulse signal as shown in FIG. The flip-flop circuit A ₃ is composed of a D-type flip-flop FF and NAND gates N ₁ and N _{2. The} output of the low frequency oscillator circuit A ₂ is divided by the D-type flip-flop FF to obtain a NAND gate.
Outputs shown in FIGS. 10B and 10C are obtained from N ₁ and N ₂ . Here, the period T _D in FIG. 10 (a) is from t _{2 of} FIG.
It corresponds to the period of t ₃ or t _{4 to} t ₅ . The high frequency oscillating circuit A ₁ generates a high frequency pulse signal as shown in FIGS. 6 (c) and 6 (d). The trigger circuit A ₄ is adapted to generate a trigger signal for turning on the bidirectional three-terminal thyristor Q ₁ every half cycle in synchronization with the output of the low frequency oscillator circuit A ₂ .

By the way, in the circuit of FIG. 8, during the starting process, the voltage Vla as shown in FIG. 11 (a) is applied to both ends of the high pressure discharge lamp DL, and the current Ila as shown in FIG. 11 (b) is applied to the high pressure discharge lamp DL. Flow to. The starting process will be described in detail with reference to FIG. 11. First, when there is no load (t _{1 to} t ₆ ) before the high pressure discharge lamp DL is lit, the voltage of the capacitor Cb is changed as shown in FIG. The voltage Vla has a rectangular wave shape in order to be charged rapidly until it becomes equal to the voltage of the capacitor Ca and to maintain the peak of the voltage until the polarity is reversed. Then, a high-voltage pulse generated by the igniter circuit G is superimposed on the voltage Vla, and the high-pressure discharge lamp DL is started by this high-voltage pulse. After t _7, the process of starting the high pressure discharge lamp DL is shown. First, at time t ₈ , a high voltage pulse is applied to the high pressure discharge lamp DL and a current Ila flows, as shown in FIG. After t _{9 when the} polarity is reversed, the current Ila does not flow, and the high voltage pulse is applied during the next negative half cycle.
It will be stopped until t ₁₀ . In FIG. 11 (b), the period Ta and the rest period Tb during which the current Ila is flowing through the high pressure discharge lamp DL.
Comparing with, Ta <Tb, and there are many pauses, and it is difficult to smoothly transition from glow discharge to arc discharge. That is, there is a shortage of energy for transitioning from glow discharge to arc discharge. Fig. 11 (b) in the normal case
The current Ila of 1 flows for only a positive side or a negative side for a period of time, and this state is repeated alternately. This is a so-called half-wave discharge phenomenon that often occurs in high pressure discharge lamps, especially metal halide lamps.
This occurs during the process of applying a high-voltage pulse to glow discharge and transition to arc discharge. This half-wave discharge is caused by the difference in the characteristics of both electrodes of the high-pressure discharge lamp DL.It is particularly likely to occur when the enclosed material inside the arc tube adheres to the electrodes and makes it difficult to discharge. The discharge lamp in which iodide is enclosed is often used. This phenomenon is unavoidable in mercury lamps and high-pressure sodium lamps, although there are some differences.

FIG. 12 shows the state of transition to arc discharge through the starting process of FIG. 11 as a waveform of the current Ila. As shown in the latter half of FIG. 12, a current Ila flows in both positive and negative directions, which means that the engine has been completely started. Due to this current, the voltage Vla gradually rises and shifts to steady lighting. However, if the power supply voltage is low or if the high-pressure discharge lamp DL is used for a long time and it is more difficult to start than in the initial state, the probability of half-wave discharge is high, and the duration is long, so
The state of FIG. 11 (b) was repeatedly performed, and the engine sometimes did not start. Such a phenomenon is likely to occur at the time of restart, and the arc discharge does not reach the glow discharge state. Therefore, the method of increasing the height, width, and number of high-voltage pulses, adding a circuit to prevent the voltage from dropping too much against fluctuations in the power supply voltage, increasing the secondary short-circuit current, and increasing the arc discharge from glow discharge When it started to shift to, the current Ila flowing at this time was doubled from the steady state. However, this increases the stress of the igniter circuit G, and even when the voltage is increased, the circuit becomes complicated, and there are drawbacks such as an increase in cost and an increase in size.

Therefore, the following method was devised as a means for smoothly starting the glow discharge to the arc discharge of the high-pressure discharge lamp without increasing the energy at the time of starting and for surely starting. That is, the operating period of the switching element operating at low frequency at the time of starting is made longer than the operating period of the switching element in the steady state, and the glow discharge is started once the discharge lamp starts to discharge using the time of the long period. Is supplied to the discharge from the arc to the arc discharge, and the switching element is reversed after the arc is discharged.

FIG. 13 shows an alternative circuit to the circuit of FIG. 9 used in the discharge lamp lighting device of FIG. 8, and a start compensation circuit B is attached to this circuit. The other circuits have the same configuration and operation as the circuit shown in FIG. In this circuit, the starting compensation circuit B is inserted between the current detection element Zs and the time constant circuit of the timer circuit tm of the low frequency oscillation circuit A ₂ . That is, when the high-pressure discharge lamp DL starts and the current Ila flows, the transistor Tr ₉ is turned on by the output of the current detection element Zs, the output Q of the D-type flip-flop FF ₁ is set to the “Low” level, and the transistor Tr ₁₀ is turned on. Then, the resistor R ₄ of the time constant circuit is short-circuited, and the low frequency oscillation circuit A ₂ is operated in the same manner as the circuit of FIG. That is, as shown in Fig. 14 (a) in the steady state.
The low-frequency oscillator circuit A ₂ is operated with the oscillation cycle of T ₁ . Then, in the period in which the current Ila is not detected, the transistor Tr ₉ is off, the output Q of the D-type flip-flop FF ₁ is at “High” level, and the transistor Tr ₁₀ is off. That is, the resistor R ₄ is connected to the time constant circuit, the time constant increases, and the oscillation cycle of the low-frequency oscillator circuit A ₂ becomes longer.

Thus, at the time of starting, since the current Ila does not flow in the high pressure discharge lamp DL as shown in the first half of FIG. 14 (b), both the transistors Tr ₉ and Tr ₁₀ in the starting compensation circuit B are turned off, The oscillation cycle of the low-frequency oscillator circuit A ₂ becomes long as shown in the first half of FIG. 14 (a). And the igniter circuit G
When the high-pressure discharge lamp DL is started by the high-voltage pulse of, the current Ila begins to flow as shown in FIG. 14 (b), and the no-load period of t _{0 to} t ₁ ends. When the current Ila begins to flow, the transistor Tr ₉ is inverted from off to on by the detection output of the current detection element Zs, as shown in FIG. 14 (c). At this time t ₁ , D
Type flip-flop FF ₁ low frequency oscillator A
_Since the output of ₂ remains at the "High" level as shown in FIG. 14 (a), the output Q of the D-type flip-flop FF ₁ is not inverted and the transistor Tr ₁₀ is in the off state. It continues for that half cycle. In other words, even if the current Ila begins to flow, the oscillation cycle is not shortened immediately, but a time sufficient for transition from the glow discharge to the arc discharge is given between t ₁ and t ₂ . Then, when the output of the low-frequency oscillator circuit A ₂ rises next (at time t ₂ ), the output Q is inverted, the transistor Tr ₁₀ is turned on as shown in FIG. 14 (d), and the low-frequency oscillator circuit A ₂ is turned on. Shifts to a steady-state oscillating operation with an oscillation cycle of T ₁ .

Although the starting performance has been improved by the above circuit configuration, for example, a discharge lamp that causes half-wave discharge (current flows only in one direction, non-conduction in the other direction) due to consumption of one electrode of the discharge lamp, etc. When the above-mentioned method is lit, the current Ila flowing through the discharge lamp becomes as shown in FIG. 15 (b), which is clear when compared with the waveform of the current Ila during normal lighting (FIG. 15 (a)). In addition, there is an inconvenience that the stress of the transistor on the conduction side increases. For this reason, it is necessary to use a transistor having a large current rating or to increase the heat dissipation plate of the transistor to reduce stress, resulting in problems such as an increase in size of the device and an increase in cost.

(Object of the Invention) The present invention has been made in view of the above points, and an object of the present invention is to reduce stress of a semiconductor switching element during half-wave discharge with a simple structure. In order to provide an electric lighting device.

DISCLOSURE OF THE INVENTION In the discharge lamp lighting device of the present invention, FIG.
As shown in the figure, during the half-wave discharge, the semiconductor switching element on the current conducting side and the semiconductor switching element on the non-conducting side radiate heat with the same heat dissipation plate, so that the stress on the semiconductor switching element on the conducting side during the half-wave discharge is reduced. It is to reduce it. By doing so, one semiconductor switching element is non-conducting at the time of half-wave discharge, so that the substantial heat dissipation area is doubled for the semiconductor switching element on the conducting side, and thermal stress is reduced. Is.

The configuration of the present invention will be described below with reference to the illustrated embodiments.
FIG. 1 shows an embodiment of the present invention, which is an example in which the transistors Tr ₅ and Tr ₈ in the circuit of FIG. ₈ are attached to the same heat sink 3a, and the transistors Tr ₆ and Tr ₇ are attached to the same heat sink 3b. . FIG. 2 shows another embodiment of the present invention, in which the transistors Tr ₅ , Tr ₈ , Tr ₆ , Tr ₇ in the circuit of FIG.
It is an example attached to. FIG. 16 shows a method of attaching the transistors Tr ₅ , Tr ₆ , Tr ₇ , Tr _{8 to} the heat sink in the conventional example. As described above, in the conventional example, since a heat sink is provided for each transistor individually, if an imbalance occurs in the conduction period of the transistors, a large thermal stress is applied to the transistor having a long conduction period. However, the heat radiation capacity of the heat radiation plate is insufficient, but the transistor having a short conduction period is hardly thermally stressed, and the heat radiation capacity of the heat radiation plate is excessive. The present invention eliminates such inconvenience, and by mounting a transistor having a long conduction period and a transistor having a short conduction period on the same heat dissipation plate, the heat dissipation capability of a transistor having a long conduction period is substantially reduced to the conventional one. This is twice that of the example.

As the lighting circuit, in addition to the full-bridge inverter circuit using four transistors as shown in FIG. 8, for example, a half-bridge inverter circuit using two transistors as shown in FIG. 3 may be used. . In the circuit of FIG. 3, the smoothing capacitors are capacitors Ca ₁ , Ca ₂
It is composed of a series circuit of transistors Tr ₁₁ , T
By switching r ₁₂ as shown in FIGS. 4 (a) and 4 (b), a rectangular-wave AC current containing high-frequency ripple is generated in the discharge lamp DL as in the circuit shown in FIG.
Is to be supplied to Also in this circuit, when controlling so that the switching control periods τ ₁ and τ ₂ of the transistors Tr ₁₁ and Tr ₁₂ are unbalanced during half-wave discharge,
The thermal stress of the transistor with the longer conduction period becomes a problem, but if transistors Tr ₁₁ and Tr ₁₂ are mounted on the same heat sink, the excess heat dissipation capability of the transistor with the shorter conduction period will be reduced. Can be used for the transistor on the long side of, and the problem of thermal stress can be solved.

(Effects of the Invention) As described above, according to the present invention, when the discharge lamp is steadily lit, the conduction is performed approximately for each synchronous period, and when the discharge lamp is not steadily lit, the conduction period is controlled to be unbalanced. In a discharge lamp lighting device having a set of semiconductor switching elements,
Since one set of switching elements whose conduction periods are unbalanced are attached to the same heat radiating element, the side which is electrically connected for a long time at the time of starting the discharge lamp or at the time of half-wave lighting at the end of its life during non-steady lighting of the lamp The semiconductor switching element and the semiconductor switching element on the short conducting side can be radiated by the same heat radiating element, and the semiconductor switching element on the long conducting side has substantially twice the heat radiating effect as in steady lighting. Since it can be expected, there is an effect that the thermal stress can be reduced.

[Brief description of drawings]

FIG. 1 is a perspective view of an essential part of an embodiment of the present invention, FIG. 2 is a perspective view of an essential part of another embodiment of the present invention, and FIG. 3 is a circuit diagram of yet another embodiment of the present invention. FIG. 4 is an operation explanatory diagram of the same as above.
FIG. 7 is a circuit diagram of a conventional example, FIGS. 6 and 7 are operation explanatory diagrams of the same as above, FIGS. 8 and 9 are circuit diagrams of other conventional examples, and FIGS. FIG. 13 is a circuit diagram of a main part of still another conventional example, FIGS. 14 and 15 are operation explanatory diagrams of the same, and FIG. 16 is a perspective view showing a conventional attachment structure to a heat sink. Tr _{5 to} Tr ₈ are transistors, 3,3a and 3b are heat sinks, DL is a discharge lamp, and A is a control circuit.

Claims (6)

[Claims] 1. A DC power supply and at least one pair of two stones.
A discharge lamp lighting device comprising a set of semiconductor switching elements, a current limiting element, and a discharge lamp that is electrically conductive from a DC power source through the current limiting element when the semiconductor switching element is conducting, The semiconductor switching element of
In the discharge lamp lighting device, the discharge lamp is controlled so that it conducts approximately at the same time during steady lighting of the discharge lamp, and during non-steady lighting, the conduction periods of the semiconductor switching elements are unbalanced. The period becomes unbalanced 1
A discharge lamp lighting device, characterized in that a pair of semiconductor switching elements are attached to the same heat dissipation element. 2. The electric current flowing through the discharge lamp when the discharge lamp is steadily lit is an alternating current.
The discharge lamp lighting device according to the item. 3. The non-steady-state lighting of the discharge lamp is characterized in that the conduction period of the one set of switching elements is controlled so as to lengthen the period in which the discharge lamp is energized. The discharge lamp lighting device according to claim 1. 4. A discharge lamp comprising at least one set of semiconductor switching elements, wherein one switching element repeats an opening / closing operation a plurality of times and the other switching element repeats an opening / closing operation a plurality of times. 2. The discharge lamp is controlled so as to be supplied with a current having a frequency lower than that of the plurality of opening / closing operations by alternately performing the current-carrying directions while being reversed. Discharge lamp lighting device. 5. The imbalance of the conduction period is caused by a difference in duration between an operation in which the one switching element repeats an opening / closing operation a plurality of times and an operation in which the other switching element repeats an opening / closing operation a plurality of times. The discharge lamp lighting device according to claim 4, wherein the discharge lamp lighting device is controlled. 6. The at least one set of semiconductor switching elements comprises a first and a second switching element in which one of them is alternately turned on and a plurality of opening / closing operations during a period in which the first switching element is on. During the period in which the third switching element that performs
A fourth switching element that performs a plurality of switching operations, the first and fourth switching elements and the second and third switching elements are connected in series, and each series circuit includes a first and second switching element. The switching element is connected to one side of the DC power source, and the third and fourth switching elements are connected in parallel to the DC power source so as to be connected to the other side of the DC power source, and the connection of the first and fourth switching elements is performed. The discharge circuit according to claim 1, wherein a series circuit of the discharge lamp and the current limiting element is connected between the point and a connection point of the second and third switching elements. Electric lighting device.