JP2010055915A - High pressure discharge lamp lighting device, light source device, and starting method of high pressure discharge lamp - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a high pressure discharge lamp lighting device in which problems of blackening, going-out, sputtering or the like due to controlling at starting up are solved without changing a conventional circuit configuration. <P>SOLUTION: In the high pressure discharge lamp lighting device provided with a chopper circuit outputting DC power, a full-bridge circuit converting the output of the chopper circuit into AC output and supplying it a high pressure discharge lamp, an ignitor circuit generating pulse voltage for starting up the high pressure discharge lamp, and a controlling part controlling the chopper circuit and the full-bridge circuit, the ignitor circuit is provided with a lighting detection means detecting an existence of lighting of the high pressure discharge lamp and outputting it to the controlling part, and the controlling part is structured so that the chopper circuit or the full-bridge circuit is controlled so that a lamp current for a first period after the lighting is detected may become the first peak value or lower and moreover the lamp current during a second period after a given period elapses may become a second peak value, and the first period is 0.5-3 seconds and the second peak value is set to become 1.2-5 times the first peak value. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to an improvement of a high pressure discharge lamp lighting device for lighting a high pressure discharge lamp, and particularly to an improvement of a starting method of the high pressure discharge lamp.

  In recent years, the high pressure discharge lamp lighting device has become smaller and lighter due to the digitization, and a combination of the step-down chopper circuit 20, the full bridge circuit 30, and the igniter circuit 40 as shown in FIG. ) High-pressure discharge lamp lighting devices that start 50 at a high frequency or a rectangular wave and then stably light at a low-frequency rectangular wave are becoming widespread. In this specification, high frequency means a frequency higher than 1 kHz, and low frequency means a frequency of 1 kHz or less.

  The operation of the circuit of FIG. 1 will be described. In the control of the PWM control circuit 28 constituting the step-down chopper circuit 20, a lamp voltage signal proportional to the lamp voltage is detected by the resistor 26, and a lamp current signal proportional to the lamp current is detected by the resistor 27. A voltage signal obtained by multiplying the lamp current signal and the lamp voltage signal by a multiplier or a voltage signal calculated by a microcomputer, and a reference voltage set in advance so that the lamp 50 can be lit at the rated lamp power at the rated lamp voltage are error amplifiers. The duty ratio of the transistor 21 is pulse width controlled so that the voltage signal obtained by multiplying the lamp current signal and the lamp voltage signal or the voltage signal calculated by the microcomputer becomes constant. As a result, the lamp 50 is turned on with desired power.

  Next, the operation of the full bridge circuit 30 that operates by receiving the DC output of the step-down chopper circuit 20 will be described. The transistors 31 and 34 and the transistors 32 and 33 are controlled by the bridge control circuit 35 so as to repeat conduction and non-conduction alternately at a predetermined frequency, whereby the direct current output of the step-down chopper circuit 20 is converted into alternating current and the lamp 50 To be supplied.

  Here, when the lamp 50 is started, in the case of high-frequency starting, the frequency controlled by the bridge control circuit 35 is increased to several tens of kHz for a certain period of time, so that the transistors 31 and 34 and the transistors 32 and 33 are connected to the middle point. The series circuit of the choke coil 36 and the capacitor 37 resonates, and a sine wave resonance voltage determined by the inductance of the choke coil 36, the capacitance of the capacitor 37, and the frequency of the bridge control circuit 35 is generated at the ends of the choke coil 36 and the capacitor 37. The high-frequency resonance voltage is also applied to the end of the lamp 50 connected in parallel to 37.

  Next, the operation of the igniter circuit 40 for starting the lamp 50 will be described. When the full-bridge circuit described above receives a high-frequency sine wave voltage generated at the end of the capacitor 37 when operating at a high frequency of several tens of kHz, a diode is connected when the connection point side of the choke coil 36 and the capacitor 37 has a positive potential. Current flows in the direction of 41, resistor 43, and capacitor 45, and capacitor 45 is charged.

  Further, when the polarity of the high frequency sine wave voltage is reversed and the connection point of the choke coil 36 and the capacitor 37 is a negative potential, a current flows in the direction of the capacitor 46, the resistor 44, and the diode 42 to charge the capacitor 46.

By repeating the above operation, the potential at the series circuit end of the capacitor 45 and the capacitor 46 gradually increases. In general, the voltage generated at the end of the capacitor 37 is determined by the resonance frequency of the resonance circuit of the choke coil 36 and the capacitor 37, and the frequency characteristics are shown in FIG. Conventionally, at the time of starting frequency characteristics near the frequency f ₅ of one-fifth of the resonance frequency fo it is used (e.g., Patent Document 1).

  More specifically, first, the operation of the full bridge circuit 30 is started at, for example, the start frequency fs, and the frequency is lowered with time. Then, the resonance voltage rises in a curve according to the frequency characteristics of FIG. As shown in FIG. 9, for example, when the discharge gap 48 breaks down at the frequency fh, that is, the voltage Vh, the voltage detection circuit 49 detects the breakdown, so that the frequency of the full bridge circuit 30 is fixed, and a constant interval is obtained. The pulse voltage is continuously generated.

  The output voltage of the chopper circuit 20 at this time receives an input DC voltage of, for example, DC 380V, and outputs a set constant voltage, for example, DC 200V. Note that the upper limit of the discharge voltage of the high-pressure discharge lamp is generally about 150 V at the highest.

  When the discharge gap 48 breaks down as described above, a pulse voltage corresponding to the boost ratio of the pulse transformer 47 is generated in the secondary winding of the pulse transformer 47 with respect to the voltage applied to the primary winding. Since the voltage is applied to the lamp 50 via the capacitor 37, the high-pressure discharge lamp 50 breaks down due to the pulse voltage, and the lamp current is subjected to arc discharge (B to E) via glow discharge (A) as shown in FIG. ).

  The lamp current value at the time of arc discharge at this time is limited to a value determined by the rectangular wave high-frequency voltage from the previous full bridge circuit 30 and determined by the inductance value of the choke coil 36, the inductance value of the pulse transformer 47 and its frequency. It becomes a thing.

  Explaining this process in a little more detail. Discharge lamps used in liquid crystal projectors and the like are generally used under air-cooling conditions, and are equipped with reflectors, so the heat capacities near the left and right electrodes are different, and air-cooling is used. Therefore, there is a difference in how the temperature near the electrode decreases when the lamp is extinguished. Then, as shown in FIG. 5A, the evaporated mercury is unevenly liquefied and adheres to the vicinity of the electrode on the side where the temperature rapidly decreases.

  In the state shown in FIG. 5A, glow discharge is started when a pulse voltage is applied to the lamp to cause dielectric breakdown. This glow discharge is affected by the biased mercury, and the electrode having the smaller heat capacity, that is, the electrode without mercury is heated first. Then, in the electrode, the temperature of the portion having a small heat capacity such as the edge of the outer coil shown in FIG. 5B first rises, and when that portion is sufficiently heated, arc discharge, which is thermionic emission, starts at that point as a spot. Is done.

  Even after the transition to arc discharge, current flows only in half-waves in the direction where the electrode on which mercury was not attached becomes the cathode, but by maintaining half-wave arc discharge, the other mercury is As the mercury on the attached electrode gradually evaporates, the temperature of the electrode also rises and current begins to flow in an asymmetrical manner (period C in FIG. 4). 4 period D).

  Then, after becoming a symmetric high-frequency current, the current is shifted to a low-frequency rectangular wave (period E in FIG. 4) current of 50 to 400 Hz, thereby preventing extinction at that time and unnecessary sputtering of the electrode. It becomes possible.

  Next, FIG. 3 shows an example of an arc tube of a discharge lamp used in a light source device for a liquid crystal projector. When the distance L from the coil end 4a on the base 3a side of the electrode 3 to the inner wall 5a of the discharge tube 5 is large, the electrode 3 embedded in the sealing portion 6 is formed. Since the vicinity of the root 3a is not warmed, the evaporation of mercury is delayed, it takes a start-up startup time, and the time required for the screen to reach a predetermined brightness after the projector device is turned on (after starting lighting) is prolonged. Is designed to be 1 mm or less.

However, in the high-pressure mercury lamp 2 whose distance is 1 mm or less, when the lighting is repeatedly performed, as described above, when starting the discharge by applying the start pulse and shifting from the glow discharge to the arc discharge, When the arc spreads instantaneously and particularly when the distance between the arc spot and the inner wall 5a of the discharge tube 5 is close as shown in FIG. 5C, the arc spreading instantaneously at the time indicated by the dotted line in FIG. Tungsten, which is an electrode constituent material, may come into contact with the inner wall 5a and cause blackening. This phenomenon occurs, for example, when a transition is made from glow discharge to arc discharge with a high peak current at the time of high-frequency starting when high-frequency lighting is performed in the vicinity of the frequency f _{5 that} is one fifth of the resonance frequency f _{0 in} FIG. I know that. This blackening causes a decrease in illuminance at an early stage, and when blackening is severe, there is a problem that a thermal load on the inner wall is increased to break the discharge tube and shorten the lamp life.

Therefore, in order to avoid the above problem, high frequency lighting is performed in the vicinity of the frequency f ₃ (f ₃ > f ₅ ), which is one third of the resonance frequency f ₀ , and the peak current at high frequency start is reduced as shown in FIG. Thus, there is a method of preventing the inner wall 5a from being contacted even if the arc spreads momentarily during the arc discharge transition. In this case, however, insufficient preheating of the electrode may occur. In other words, when the temperature of the electrode on the anode side during half-wave lighting does not rise, and when it shifts to low-frequency rectangular wave lighting after a certain time, the electrode becomes the cathode, the time indicated by the dotted line in FIG. In some cases, arc discharge cannot be performed, and sputtering may occur more than necessary, or the lamp may disappear.

  Further, in the method of reducing the peak current at the time of high frequency lighting as described above, the temperature of both electrodes is sufficiently increased at a high frequency with a long high frequency lighting time and a small current, and then a transition to low frequency rectangular wave lighting is performed. There is also a method. However, in this case, considering the time of restart (starting when the arc tube is not cooled), since the pressure in the arc tube is high from the beginning, the high-pressure lighting time is increased to increase the pressure in the arc tube. The standing wave in the lower part grows and may cause flickering or disappearance due to the acoustic resonance phenomenon.

As a countermeasure for the above problem, as shown in FIGS. 7 and 8, the peak current at the time of high-frequency lighting is reduced for a certain period (50 ms to 500 ms) starting from f ₃ s in period B1, and the process shifts to f _{5 in} subsequent period B2. There is also a method of lighting with a normal high-frequency starting current (for example, Patent Document 2).

JP 2004-127656 A JP 2008-108670 A

However, in the case of Patent Document 2 (FIG. 7), if when the lamp is lit in the period B2 of f ₅ not lit in the period f ₃ B1, it may also cause blackening as described above.

  Therefore, it is detected that the lamp is lit, and a high frequency starting current that is small so as not to be blackened for a certain period after lighting is supplied to the lamp. It needs to flow through the lamp. However, even if a pulse is generated by the high frequency operation and the lamp is turned on, the impedance of the choke coil 36 becomes large because the frequency is high during the high frequency operation period. For this reason, almost no current flows through the lamp while the output of the chopper circuit unit 20 is maintained at the upper limit value DC200V. Therefore, it is impossible to detect whether the lamp is lit by the voltage detector A in FIG. 1 that detects the lamp voltage during normal low-frequency operation or the current detection resistor 27 in FIG. As another method, an auxiliary winding or the like is wound around the choke coil 36, and the voltage at that portion can be detected to determine whether or not the lamp has been lit. Neither the surface nor the space in designing the board is appropriate.

  Therefore, the present invention solves the above problem, that is, the blackening that can occur in the period B in FIG. 4 and the period B2 in FIG. 7 and the disappearance that can occur in the transition from the period C to E in FIG. The technical problem is to solve the problem in terms of control without changing the (hard part), and to improve the startability of the lamp and extend its life.

  A first aspect of the present invention is a chopper circuit that outputs DC power, a full-bridge circuit that converts the output of the chopper circuit into an AC output and supplies it to a high-pressure discharge lamp, and generates a pulse voltage for starting the high-pressure discharge lamp. In a high pressure discharge lamp lighting device having a control unit for controlling an igniter circuit and a chopper circuit and a full bridge circuit, the igniter circuit includes lighting detection means for detecting the presence or absence of lighting of the high pressure discharge lamp and outputting it to the control unit The second current so that the lamp current in the first period after the lighting detection becomes equal to or less than the first peak value and the lamp current in the second period after the predetermined period has elapsed is larger than the first peak value. The control unit is configured to control the chopper circuit or the full bridge circuit so as to have a peak value, the first period is longer than 0.5 seconds and not longer than 3 seconds, and the second peak value is A high pressure discharge lamp lighting device less than 5-fold 1.2-fold or more of the peak value.

  Here, the resonance circuit may include an inductor connected in series to the high-pressure discharge lamp, and the control unit may lower the drive frequency of the full bridge circuit when the first period is shifted to the second period.

  In addition, the control unit may increase the output voltage of the chopper circuit when shifting from the first period to the second period.

  Further, the full bridge circuit includes a resonance circuit that generates a resonance voltage by a switching operation of the full bridge, and the igniter circuit is a charging circuit that charges the resonance voltage, and a discharge gap that is short-circuited when the voltage generated in the charging circuit exceeds a predetermined value. The primary winding is composed of a discharge transformer and the secondary winding is connected to a high-pressure discharge lamp, and the lighting detection means is composed of a voltage detection circuit for detecting the voltage of the charging circuit.

  According to a second aspect of the present invention, there is provided a light source device comprising the high pressure discharge lamp lighting device, the high pressure discharge lamp, the reflector to which the high pressure discharge lamp is attached, and a housing containing at least the high pressure discharge lamp lighting device. is there.

  A third aspect of the present invention is a method for starting a high pressure discharge lamp, wherein (A) a step of applying a start pulse to the high pressure discharge lamp by an igniter circuit, and (B) lighting of the high pressure discharge lamp in the igniter circuit. A step of detecting the start, (C) a step of energizing the high-pressure discharge lamp with a high-frequency current equal to or lower than the first peak value by the AC output circuit in a predetermined period after the start of lighting is detected, and (D) elapse of the predetermined period Later, the method includes a step of energizing the high-pressure discharge lamp with a high-frequency current having a second peak value higher than the first peak value by an AC output circuit, wherein the first period is longer than 0.5 seconds and not longer than 3 seconds, In this method, the second peak value is not less than 1.2 times and not more than 5 times the first peak value.

  Here, an inductor may be connected in series to the high-pressure discharge lamp, and step (D) may include a step of reducing the drive frequency of the AC output circuit.

  Further, step (D) may include a step of increasing the output voltage of the AC output circuit.

  According to the present invention, it is possible to solve problems such as blackening, disappearance, and sputtering caused by control at the time of starting without changing the conventional circuit configuration, and to improve the startability and extend the life of the lamp.

  The present invention detects the start of lamp discharge, limits the peak current of half-wave lighting at the time of arc discharge transition for a predetermined time, and then shifts to a high-frequency current value of a design value that assumes arc discharge due to full-wave current. is there. In particular, the above-described sequence is achieved by reliably performing the measurement for the predetermined period without changing the conventional circuit configuration (portion shown in the circuit diagram).

  Since the circuit configuration of the high pressure discharge lamp lighting device in the present invention is the same as that of FIG. 1 as a conventional example, and only the control is different, the control will be described below.

  Similar to the conventional example, the choke coil 36 and the capacitor 37 resonate by alternately turning on and off the transistors 31 and 34 transistors 32 and 33 of the full bridge circuit at a high frequency of several tens of kHz, and the choke coil 36 and the capacitor 37 are connected. When the point side has a positive potential, current flows in the direction of the diode 41, the resistor 43, and the capacitor 45, and the capacitor 45 is charged. When the polarity of the high-frequency sine wave voltage is reversed and the connection point between the choke coil 36 and the capacitor 37 is a negative potential, a current flows in the direction of the capacitor 46, the resistor 44, and the diode 42, and the capacitor 46 is charged. By repeating the above operation, the potential at the series circuit end of the capacitor 45 and the capacitor 46 gradually increases. When the end of the series circuit of the capacitors 45 and 46 exceeds the breakover voltage of the discharge gap 48, the discharge gap 48 causes dielectric breakdown, and the electric charge accumulated in the capacitors 45 and 46 is discharged to the primary side of the pulse transformer 47 all at once. A pulse voltage corresponding to the turn ratio is generated on the secondary side.

  Then, the lamp 50 to which the pulse voltage is applied causes dielectric breakdown, and glow discharge is started as shown in FIG. 2A. Then, similarly to the conventional example, it is influenced by the biased mercury, and the temperature of the edge of the outer coil having a small heat capacity becomes high as shown in FIG. 5B, and the transition to arc discharge starts from that portion. . At this time, although depending on the shape of the electrode and the arc tube, by setting the frequency at which the peak value of the arc discharge current half-wave discharged becomes, for example, 2A or less as shown in FIG. When the inner coil edge becomes an arc spot and the transition from glow discharge to arc discharge occurs, even if the arc spreads instantaneously, the arc does not become large enough to contact the inner wall 5a in FIG. No blackening occurs inside the discharge tube.

  In addition, if the current value is, for example, 2A peak in the half-wave discharge, the temperature rise of the electrode becomes insufficient and may disappear. Therefore, the period during which the half-wave discharge is limited to the 2 A peak (zero-peak) current is limited to, for example, only one second (period B in FIG. 2) after detecting that the lamp is lit, A current that can be sufficiently preheated, for example, 6 A peak (peak-peak) for 2 seconds is set (period D). By doing so, the temperature of the electrode on the side to which mercury is not attached sufficiently rises during the high-frequency lighting period, and a smooth transition to arc discharge becomes possible. Even if the half-wave state continues even after 1 second, since the electrode is heated in a state of a small high-frequency starting current, there is a merit that it is easy to shift to a full wave when shifting to a large current.

  The period B is suitably longer than 0.5 seconds and not longer than 3 seconds, and the peak current value (zero-peak) is not less than 1A and not more than 4A. It is appropriate that the period D is 0.5 seconds or more and 3 seconds or less and the peak current value is 1.2 times or more and 5 times or less the current value of the period B. In particular, the total time of the period B and the period D is preferably about 3 seconds (± 0.5 seconds), but is not limited thereto.

  Here, as means for detecting that the lamp has been turned on, means for detecting that the lamp has been turned on by the conventional circuit of FIG. 1 without adding components will be described.

  As described above, when the lamp 50 is not lit even when the pulse voltage is generated, the capacitors 45 and 46 are repeatedly charged and discharged. That is, the discharge gap end voltage repeats dielectric breakdown / non-dielectric breakdown as shown in FIG. 9, and the voltage constantly fluctuates.

  However, when a pulse voltage is generated to cause dielectric breakdown and the lamp 50 is lit, a resonance voltage is applied to the lamp. Therefore, no electric charge is accumulated in the capacitors 45 and 46, so that the potential does not rise, that is, the discharge gap end voltage does not rise as shown in FIG. The voltage detection circuit 49 detects that the voltage does not increase for a certain period of time, and it can be reliably determined whether or not the lamp 50 is lit.

  If it is possible to accurately detect that the lamp 50 is lit, the high-frequency current after lighting can be controlled appropriately. For example, if the peak value of the high-frequency current is switched to two stages in order to make a smooth arc discharge, (1) the frequency of the full bridge circuit unit 30 after detecting the lighting of the lamp by the above method. (2) Switch the output voltage of the DC power supply circuit unit 20 in FIG. 1 (the circuit component may need to have a higher withstand voltage, but the number of components and the component arrangement may be changed.) (There will be no change). Even if they are combined or switched in multiple stages instead of two stages, the same or higher effect can be obtained.

  Further, if the purpose of the present invention is not to change the configuration of the conventional circuit, but only to increase the lamp current during the transition from the period B to the period D, the above (1) or In addition to (2), means for connecting an additional inductor in series to the resonant circuit, connecting a switch in parallel to the additional inductor, and changing the resonance point of the resonant circuit by opening and closing the switch (current increase when the switch is closed) is also conceivable. This is an effective means when there is room for mounting additional inductors and switches in the board space.

  In the above-described embodiment, a high pressure discharge lamp lighting device that prevents blackening, extinction, and sputtering due to the starting operation is shown. A light source device as an application using the high pressure discharge lamp lighting device is shown in FIG. In FIG. 10, 61 is the high pressure discharge lamp lighting device of the embodiment described above, 62 is a reflector to which the high pressure discharge lamp 50 is attached, 63 is a housing containing the high pressure discharge lamp lighting device 61, the high pressure discharge lamp 50 and the reflector 62. Is the body. In addition, the figure is a schematic illustration of the embodiment, and the dimensions, arrangement, and the like are not as illustrated. Then, a projector is configured by appropriately arranging a video system member or the like (not shown) in the housing 63.

  Thereby, it is possible to obtain a light source device with good startability, a long lamp replacement cycle, and high reliability.

It is a circuit diagram of the conventional high pressure discharge lamp lighting device. It is a figure which shows the lamp current waveform at the time of starting of the discharge lamp lighting device in this invention. It is a figure which shows the arc tube of a general high pressure discharge lamp. It is a figure which shows the lamp current waveform example at the time of starting of the conventional high pressure discharge lamp lighting device. It is a figure explaining the mercury and arc spot of the arc tube of a high pressure discharge lamp. It is a figure explaining the mercury and arc spot of the arc tube of a high pressure discharge lamp. It is a figure explaining the mercury and arc spot of the arc tube of a high pressure discharge lamp. It is a figure which shows the lamp current waveform at the time of starting of the conventional discharge lamp lighting device. It is a figure which shows the lamp current waveform at the time of starting of the conventional discharge lamp lighting device. It is a characteristic figure of the resonance circuit explaining operation of a high pressure discharge lamp lighting device. It is a figure which shows each operation | movement waveform which shows the operation | movement in the high frequency operation period of a high pressure discharge lamp lighting device. It is a figure which shows the light source device of this invention.

Explanation of symbols

10: DC power supply 20: Chopper circuit 21: Transistor 22: Diode 23: Choke coil 24: Capacitors 25, 26, 27: Resistor 28: PWM control circuit 30: Full bridge circuits 31, 32, 33, 34: Transistor 35: Bridge Control circuit 36: choke coil 37: capacitor 40: igniter circuit 41, 42: diode 43, 44: resistor 45, 46: capacitor 47: pulse transformer 48: discharge gap 49: voltage detection circuit 50: high pressure discharge lamp (lamp)
61: High pressure discharge lamp lighting device 62: Reflector 63: Housing

Claims (8)

A chopper circuit for outputting DC power, a full bridge circuit for converting the output of the chopper circuit into an AC output and supplying the high pressure discharge lamp, an igniter circuit for generating a pulse voltage for starting the high pressure discharge lamp, and the In a high pressure discharge lamp lighting device including a chopper circuit and a control unit that controls the full bridge circuit, the igniter circuit includes a lighting detection unit that detects whether or not the high pressure discharge lamp is lit and outputs it to the control unit,
A second current in which the lamp current in the first period after the lighting is detected is less than or equal to the first peak value, and the lamp current in the second period after the predetermined period has elapsed is greater than the first peak value. The control unit is configured to control the chopper circuit or the full bridge circuit so that the first period is longer than 0.5 seconds and not longer than 3 seconds. A high pressure discharge lamp lighting device having a peak value of 1.2 times or more and 5 times or less of the first peak value. The high pressure discharge lamp lighting device according to claim 1, wherein the resonance circuit includes an inductor connected in series to the high pressure discharge lamp,
The high pressure discharge lamp lighting device configured such that the control unit lowers the drive frequency of the full bridge circuit when the first period is shifted to the second period.   2. The high pressure discharge lamp lighting device according to claim 1, wherein the control unit increases the output voltage of the chopper circuit when shifting from the first period to the second period. apparatus. The high pressure discharge lamp lighting device according to claim 1, wherein the full bridge circuit includes a resonance circuit that generates a resonance voltage by a switching operation of the full bridge,
The igniter circuit charges the resonance voltage, a discharge gap that is short-circuited when a voltage generated in the charging circuit exceeds a predetermined value, and a primary winding in the discharge gap and a secondary winding in the high-voltage discharge. It consists of a pulse transformer connected to an electric light,
A high pressure discharge lamp lighting device comprising a voltage detection circuit in which the lighting detection means detects the voltage of the charging circuit.   A light source device comprising: the high pressure discharge lamp lighting device according to claim 1; a high pressure discharge lamp; a reflector to which the high pressure discharge lamp is attached; and a housing containing at least the high pressure discharge lamp lighting device. A method for starting a high pressure discharge lamp, comprising:
(A) applying a starting pulse to the high pressure discharge lamp by an igniter circuit;
(B) detecting the start of lighting of the high-pressure discharge lamp in the igniter circuit;
(C) energizing the high-pressure discharge lamp with a high-frequency current equal to or lower than a first peak value by an AC output circuit during a predetermined period after the start of lighting is detected; and (D) after the predetermined period has elapsed, Energizing the high-pressure discharge lamp with a high-frequency current having a second peak value higher than the first peak value by an output circuit;
The method, wherein the first period is longer than 0.5 seconds and not longer than 3 seconds, and the second peak value is not less than 1.2 times and not more than 5 times the first peak value.   The method according to claim 6, wherein an inductor is connected in series to the high-pressure discharge lamp, and the step (D) includes a step of reducing a driving frequency of the AC output circuit.   The method according to claim 6, wherein the step (D) includes a step of increasing an output voltage of the AC output circuit.

Images