JP2007335306A - Lighting device and lighting method of high-luminance discharge lamp - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To enable to stably light lamps irrespective of their individual differences by obtaining relations between lamp-state parameters and stable lighting frequencies for a whole period from a start of lighting to a normal state of a high-luminance discharge lamp. <P>SOLUTION: By carrying out trial lighting by changing lighting frequencies so as to avoid unstable states based on signals showing whether the lamp is stably lit, a relation between the lamp-state parameters and stable lighting frequencies are obtained. At normal states, lighting is carried out at the stable lighting frequency with the use of the relation during the normal state. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a lighting device and a lighting method for a high-intensity discharge lamp that stably illuminates a high-intensity discharge lamp (HID lamp: High Intensity Discharge Lamp), such as a metal halide lamp used as a light source for an automobile headlight or the like Is.

There are attempts to drive the HID lamp at a high frequency for the purpose of downsizing the HID lamp drive circuit and reducing the cost. However, when the HID lamp is lit at a high frequency of several tens of kHz to several hundreds of kHz, a so-called acoustic resonance phenomenon, which is a resonance phenomenon caused by sound waves in the lamp, occurs, the lamp becomes unstable, the arc is bent, and light is emitted It may flicker or disappear. Therefore, a method of driving at a frequency (stable lighting frequency) that does not cause arc fluctuation due to acoustic resonance has been proposed.
Since the stable lighting frequency changes depending on the lamp lighting state from the lighting start to the steady state, for example, in a conventional HID lamp lighting device, the lighting frequency is gradually increased according to the lamp impedance from the lighting start to the steady state. A method of decreasing is adopted (see, for example, Patent Document 1).

Japanese Patent Laid-Open No. 9-63783 (paragraph numbers [0019] to [0024], FIG. 1)

  However, in the conventional high-intensity discharge lamp lighting device, the relationship between the lamp impedance and the stable lighting frequency varies depending on individual differences between the lamps. Therefore, if the value of the lamp impedance and the lighting frequency is uniquely determined, depending on the lamp, Sometimes became anxious.

  The present invention has been made to solve the above-described problems. The relationship between the lamp state parameters such as the lamp impedance and the stable lighting frequency from the start of lighting of the high-intensity discharge lamp to the whole period until the steady state is obtained. The purpose is to obtain it for each lamp and to light it stably regardless of individual lamp differences.

According to the present invention, in a lighting device for a high-intensity discharge lamp whose stable lighting frequency varies depending on an internal state, it is determined whether the high-intensity discharge lamp is lit stably, in a stable state or in an unstable state. Stable state determination means for outputting a stability determination signal indicating whether there is,
Based on the stability determination signal, trial lighting of the high-intensity discharge lamp is performed while changing the lighting frequency so as to avoid an unstable state, and the lamp state parameter indicating the lighting state at that time and the locus of the lighting frequency are stored. Thus, there is provided stable lighting frequency acquisition means for acquiring stable lighting frequency information, and driving means for driving the high-intensity discharge lamp with a lighting frequency determined based on the stable lighting frequency information.

  Further, the present invention provides a lighting method for a high-intensity discharge lamp in which a stable lighting frequency band changes depending on an internal state, based on a stability determination signal indicating whether or not the high-intensity discharge lamp is stably lit. The test lighting is performed while changing the lighting frequency for the high-intensity discharge lamp so as to avoid the lamp, the lamp state parameter indicating the lighting state of the entire period from the start of lighting of the high-intensity discharge lamp to the steady state, and the stable lighting frequency This includes an initialization mode in which stable lighting frequency information indicating the relationship is acquired for each high-intensity discharge lamp and a normal lighting mode in which normal lighting is performed at a stable lighting frequency based on the stable lighting frequency information.

  According to the present invention, since the relationship between the lamp state parameter and the stable lighting frequency can be obtained for each lamp, the entire period from the start of lighting of the high-intensity discharge lamp to the steady state regardless of the individual difference of the lamp. It can be lit stably without causing the occurrence of flickering of light.

Embodiment 1 FIG.
FIG. 1 is a block diagram showing a lighting device for a high-intensity discharge lamp 1 according to Embodiment 1 of the present invention.
The lighting device includes a power supply circuit 2, a drive circuit 3, a lamp current detection circuit 4, a lamp voltage detection circuit 5, and a control circuit 6.
The drive circuit 3 includes an inverter, a resonance circuit, and a high voltage generation circuit (lamp starting circuit). In accordance with the control signal sent from the control circuit 6, the DC voltage supplied from the power supply circuit 2 is converted into a high-frequency AC voltage in the range of several tens of kHz to several hundreds of kHz, and the lamp 1 is driven. At the time of starting the lamp, the drive circuit 3 applies a high voltage (several kV to about 20 kV or so) necessary for starting the lamp (starting lighting) to the lamp.

  FIG. 2 is a circuit diagram showing a specific configuration of the drive circuit 3. The drive circuit 3 includes a full bridge inverter circuit 31 using a switching element such as an FET, a gate drive circuit 32 that drives the gate of the FET, a transformer 33, and an LC configured by a series inductor Ls, a parallel capacitor Cp, and a series capacitor Cs. A resonance circuit 34 and a high voltage generation circuit (lamp starting circuit) 35 are included.

  The inverter circuit 31 can be a full bridge circuit, a half bridge circuit, a push-pull circuit, or the like. The transformer 33 is used to insulate the inverter circuit 31 from the lamp 1 and to convert the power supply voltage into a voltage suitable for driving the lamp 1, but this can be omitted.

  The LC resonance circuit 34 functions as an impedance converter that shapes the voltage and current to be applied to the lamp 1 to approximate a sine wave and varies the voltage and current to be supplied according to the impedance of the lamp 1.

  The high voltage generation circuit 35 charges the series capacitor Cs with a high voltage of about several kV to 30 kV necessary for starting the lamp 1. As the high voltage generation circuit 35, for example, a booster circuit using a Cockcroft-Walton circuit can be used. For the high voltage generation circuit 35, another method such as a method of applying a pulse voltage of several tens of kV using an igniter transformer may be used.

  The lamp current detection circuit 4 and the lamp voltage detection circuit 5 detect the current flowing through the lamp 1 and the voltage applied to both ends of the lamp. The lamp voltage detection circuit 5 can be constituted by a voltage dividing circuit using a capacitor or a resistor and a rectifier circuit. A voltage indicating a lamp voltage is obtained by converting a high voltage applied to the lamp 1 to a relatively low voltage by the voltage dividing circuit and further converting an AC waveform applied to the lamp 1 to a DC by a rectifier circuit. . The lamp current detection circuit 4 can be realized using a current transformer or a current detection resistor.

The control circuit 6 is constituted by a microprocessor, an A / D converter, a memory, and other digital circuits and analog circuits. The control circuit 6 sends a control signal to the drive circuit 3 to control the drive frequency and output pulse width in the drive circuit.
The control circuit 6 includes a power control block 61, a stable state determination block 62 that is a stable state determination unit, a drive frequency control block 63, a stable lighting frequency storage block 64, and a stable lighting frequency acquisition block 65 that is a stable lighting frequency acquisition unit. And functional blocks such as a control signal generation block 66. These functional blocks do not necessarily correspond one-to-one with the hardware. For example, the microprocessor is related to a plurality of functional blocks such as the power control block 61 and the lighting frequency control block 63.

  The power control block 61 feeds back the detection result of the lamp current and the lamp voltage, and controls the power supplied to the lamp 1 to a predetermined value. Here, so-called PWM control is performed in which the lamp power is controlled by varying the output pulse width of the inverter serving as the driving means.

  The stable lighting frequency storage block 64 stores stable lighting frequency information, which is a relationship between the lamp state parameter and the stable lighting frequency of the lamp 1. Here, the lamp state parameter is any parameter indicating the lighting state of the lamp 1, and here, an elapsed time from the start of the lamp is used as the lamp state parameter. The elapsed time from the start of the lamp can be measured using a timer built in the microprocessor. Here, the lamp start timing can be detected from a change in the lamp voltage or lamp current. The stable lighting frequency storage block 64 can be realized by using a rewritable nonvolatile memory such as a flash memory, an EEPROM, or a battery-backed RAM.

  The drive frequency control block 63 refers to the stable lighting frequency information from the stable lighting frequency storage block 64 based on the lamp state parameter (lamp voltage) and controls the driving frequency.

  The stable state determination block 62 determines whether the lamp 1 is in a stable state or an unstable state from the lamp voltage by detecting the fluctuation of the lamp voltage that occurs when the lamp 1 is in an unstable state. Output a signal. The stable state determination block 62 can be realized by using a lamp voltage as an input and using a band-pass filter, a smoothing circuit, and a comparator. When the discharge in the lamp 1 becomes unstable due to acoustic resonance, the lamp voltage fluctuates at a frequency of about 10 Hz to several hundreds Hz. Therefore, this frequency band is extracted by a band pass filter, and the intensity of the frequency band is determined by a smoothing circuit. A signal indicating the degree of instability can be obtained by converting into an intensity signal. Further, by using a comparator, a signal indicating in binary whether the lamp is unstable or stable can be obtained.

  The stable lighting frequency acquisition block 65 acquires the relationship between the lamp state parameter and the stable lighting frequency by a method described later using the stability determination signal, and stores the information in the stable lighting frequency storage block 64.

  The control signal generation block 66 generates a control signal for controlling the drive circuit 3 based on information such as the drive frequency determined by the drive frequency control block 63 and the pulse width determined by the power control block 61.

  When the lamp 1 is lit at a high frequency of several tens of kHz to several hundreds of kHz, the discharge may become unstable due to an acoustic resonance phenomenon depending on the lighting frequency. However, there are several frequency bands that can be lit stably, called stable lighting frequency bands or stable windows, in the frequency band of several tens of kHz to several hundreds of kHz where the acoustic resonance phenomenon occurs.

  This stable lighting frequency band can be considered as a frequency at which no standing wave is generated in the HID lamp, as one interpretation. Alternatively, as another interpretation, it can be considered as a frequency band in which a standing wave where the discharge is stabilized appears among various modes of the standing wave that may exist in the lamp. In the latter case, it is considered that the arc shape viewed from the side surface of the lamp is not a circular arc shape that is normally seen, but a linear shape.

The frequency (acoustic resonance frequency) at which a specific standing wave is caused by the acoustic resonance phenomenon depends on the lamp shape, the temperature in the lamp, and the average molecular weight of the gas sealed in the lamp. The temperature inside the lamp changes depending on the lighting time from start-up, and the average molecular weight also changes due to evaporation of substances such as metal halide and mercury enclosed in the lamp due to the temperature rise of the lamp inner wall. The frequency and the stable lighting frequency band gradually change according to the time from the start of lighting.
In addition, the acoustic resonance frequency and the stable lighting frequency band also change due to individual differences due to lamp manufacturers, product lots, manufacturing variations, etc., or changes over time.

FIG. 3 schematically shows the relationship between the elapsed time from the start of lighting and the stable lighting frequency band of a 35 W metal halide lamp used for automobile headlights. Experiments have shown that the stable lighting frequency band rises for a while after lighting, then turns down, and finally becomes stable in a certain frequency band.
Since the temperature in the lamp rises rapidly for a while after the lighting, the stable lighting frequency band gradually rises. Thereafter, the stable lighting frequency band is considered to decrease as the metal halide or mercury enclosed in the lamp evaporates.
In addition, there are a plurality of stable lighting frequency bands, and the widest stable window is, for example, a so-called mercury-free HID bulb that does not contain mercury in the tube. Finally, it becomes constant at about 120 kHz.

Next, the operation will be described.
The basic operation of the present invention is performed by an initialization mode in which the relationship between the lamp state parameter and the stable lighting frequency is acquired and stored after the lamp is installed or replaced, and then the normal lighting mode used during normal lighting.

First, the normal lighting mode will be described. FIG. 4 shows a flowchart of the control procedure in the normal lighting mode.
First, a high voltage generating circuit (lamp starting circuit) 35 in the drive circuit 3 applies a high voltage of several kV to several tens of kV to the lamp, and lighting is started (S1). Thereafter, the driving circuit 3 applies a high frequency of several tens of kHz to several hundreds of kHz to the lamp to continue lighting. At this time, the lamp state parameters such as the elapsed time from the start of the lamp are acquired (S2), and according to the internal state of the lamp according to the relationship between the lamp state parameter and the stable lighting frequency stored in advance, that is, the stable lighting frequency information The stable lighting frequency is read out, and the stable lighting frequency is stored and set (S3, S4). When flickering occurs during the lighting of the lamp in the lighting process due to a change over time or the like, the set stable lighting frequency is corrected according to the stable lighting frequency information (S5, S6). Thereby, it is possible to stably light the lamp from the start of lamp lighting to the steady state.

Next, the initialization mode will be described.
In the initialization mode, lighting is performed along one stable lighting frequency band using the stability determination signal, and the locus is stored. Specifically, the operation starts at a predetermined frequency (a value that is considered to be slightly lower than the predicted value of the initial value of the widest stable lighting frequency band) immediately after the start of lighting, and thereafter, the stability determination signal is referred to On the other hand, lighting is performed while changing the lighting frequency as follows between the stable state and the unstable state.
When stable: The lighting frequency is not changed.
In an unstable state: The lighting frequency is increased or decreased at a predetermined rate.

  As shown in FIG. 3, the stable lighting frequency band has a characteristic that it rises at the beginning of lighting and then turns down, so that a predetermined lamp state parameter (time from the start of lighting) is set as an operation switching point in advance. Until the switching point is reached, the lighting frequency in the unstable state is increased at a predetermined rate, and after passing the operation switching point, the lighting frequency in the unstable state is decreased at a predetermined rate. . FIG. 5 shows a time-frequency trajectory at this time.

  Thus, since the lighting frequency changes so as to avoid this in an unstable state, lighting can be performed while the lighting frequency follows the stable lighting frequency. At this time, by sequentially storing the lamp state parameter (here, the time from the start of lighting) and the lighting frequency at that time, the relationship between the lamp state parameter and the stable lighting frequency, that is, stable lighting frequency information is accumulated. Can do.

  As described above, the relationship between the lamp state parameter and the stable lighting frequency varies depending on the individual difference between the lamps, but in the present invention, this relationship is obtained for each lamp by performing lighting in the initialization mode as described above. Since it is automatically acquired and then lit using the acquired relationship in the normal lighting mode, it can be lit stably regardless of individual differences in lamps.

FIG. 6 shows a flowchart in the case where the above control is performed. At each predetermined control cycle (Δt), the stability determination signal is used to identify whether the state is stable or unstable (S11-S14). If stable, the lighting frequency is not changed. The lighting frequency is changed by the value (Δf) (S15-S17), and the lamp state parameter and the lighting frequency at that time are stored (S18).
That is, when the unstable state continues, the lighting frequency changes by Δf for every Δt, which is almost equivalent to the change of the lighting frequency gradually with time at the slope of Δf / Δt. FIG. 7 shows a time-frequency trajectory at this time. If the lighting frequency is written in the memory of the stable lighting frequency storage block 64 in FIG. 1 for each Δt, the relationship between the time from the start of lighting and the frequency can be stored as stable lighting frequency information. The lighting frequency may not necessarily be stored every Δt, but may be stored at an interval of n × Δt (n is an integer of 2 or more), or from the start of lighting regardless of the frequency control cycle Δt. It is also possible to memorize the relationship between time and frequency.

As is apparent from FIGS. 5 and 7, lighting in the initialization mode does not always allow the lamp to be lit in a stable state, but lighting is performed along the boundary between the stable state and the unstable state. Done. Therefore, if the lamp is turned on using only the initialization mode, instability of discharge cannot be completely avoided. In the present invention, the initialization mode is used only for obtaining stable lighting frequency information in the first lighting after lamp replacement or the like, and is actually performed by the normal lighting mode based on the stable lighting frequency obtained by the initialization mode. By performing lighting, stable lighting is realized from the lighting start to the steady state.
Furthermore, in the normal lighting mode, if the stable lighting frequency information is sequentially updated so as to approach the actual stable lighting frequency band, more stable lighting can be performed as lighting is repeated.

  In the above description, the elapsed time from the start of lighting is used as the lamp state parameter, but the lamp voltage may be used as the lamp state parameter. That is, the lamp voltage is measured for each control cycle, and the relationship between the lamp voltage and the lighting frequency is accumulated as stable lighting frequency information. In the normal lighting mode, the lamp voltage is measured and the frequency corresponding to the voltage is measured. Drive.

When the lamp power is controlled to a predetermined value, the lamp voltage changes in correlation with the temperature inside the lamp, the partial pressure of the substance enclosed in the lamp, such as xenon and metal halide, and so on. It can be used as a lamp state parameter indicating the lighting state.
If the lamp state at the start of lighting, for example, the time from when it was last turned off until it is turned on, or the environmental conditions, etc., the internal state of the lamp may be different even if the elapsed time from the start of lighting is the same However, the lamp voltage is considered to accurately reflect the internal state of the lamp regardless of these conditions. On the other hand, when the lamp characteristics are close to the constant voltage characteristics, it is difficult to grasp the internal state from the lamp voltage, and both the case where the elapsed time is used and the case where the lamp voltage is used are both pros and cons.

  Therefore, it is possible to use a combination of two lamp state parameters, such as determining the initial state of the lamp by referring to the lamp voltage at the beginning of lighting, and then determining the lighting frequency according to the elapsed time after the start of lighting.

  In addition to the elapsed time from lighting and the lamp voltage, any parameter that reflects the internal state of the lamp, such as lamp current, lamp impedance, and lamp temperature, can be used as the lamp state parameter. Also, a combination of them can be used as the lamp state parameter.

Embodiment 2. FIG.
In the first embodiment, the lighting frequency is fixed without being changed in the stable state. However, the lighting frequency is not necessarily fixed, and may be gradually changed with time at a predetermined slope df / dt. That is, when it is known in advance how much the rate of change of the stable lighting frequency with respect to time is, it may be changed at a rate smaller than the minimum value of the range.
Further, in an unstable state, it may be changed at a larger rate, that is, at a rate larger than the maximum value of the rate of change of the stable lighting frequency with respect to time.
Since the time change rate of the stable lighting frequency changes according to the elapsed time from the start of lighting, it is preferable to change the slope for each time region.

  An example is shown in FIG. In FIG. 8, the time region is a region where the stable lighting frequency at the beginning of lighting suddenly increases (time region A), the stable lighting frequency is near the maximum (time region B), and the region where the stable lighting frequency gradually decreases (time region). C) Roughly divided into four regions of a steady state (time region D) where the stable lighting frequency is constant, as shown in FIG. 9 depending on whether each region is stable or unstable. Change the lighting frequency.

  In this way, by making the rate of changing the lighting frequency close to the maximum / minimum value of the actual rate of change of the stable lighting frequency, the stable lighting frequency can be obtained more accurately without leaving much of the stable lighting frequency. it can.

Embodiment 3 FIG.
In the first and second embodiments, the lighting frequency is fixed or gently changed in the stable state, and the lighting frequency is changed relatively steeply in the unstable state. According to this operation, as shown in FIG. 5, in the process of increasing the stable lighting frequency band at the beginning of discharge, tracing is performed along the lower end of the stable lighting frequency band, and then in the process of decreasing the stable lighting frequency band. Is traced along the upper end of the stable lighting frequency band.

  At this time, the operation switching point is set so as to coincide as much as possible with the point where the stable lighting frequency band turns from rising to lowering (the point where the stable lighting frequency becomes maximum). However, since the point at which the stable lighting frequency is maximized is actually different for each lamp, the average characteristic of the lamp is determined in the vicinity of the point at which the stable lighting frequency is maximized. If this operation switching point is close to the point at which the stable lighting frequency is actually maximized, there is no problem. However, if the operation switching point is greatly deviated, the following problem may occur.

That is, as shown in FIG. 10, when the operation is switched too early, the frequency changes so as to deviate from the stable lighting frequency band in the unstable state, and the stable lighting frequency is far away from the stable lighting frequency band. You may not be able to return to the belt.
Therefore, in the third embodiment, the lower end or the upper end of the stable lighting frequency band is always traced from the start of discharge to the steady state. Specifically, the lighting frequency is changed as shown in FIGS.

Here, “slowly increase” and “slowly decrease” mean that the lighting frequency is increased or decreased at a change rate smaller than the minimum value of the change rate of the stable lighting frequency with respect to time.
“Steeply increasing” and “steeply decreasing” mean that the lighting frequency is increased or decreased at a change rate larger than the maximum value of the change rate of the stable lighting frequency with respect to time. 11 and 12, (a) is referred to as rising lower end detection, (b) is referred to as falling lower end detection, (c) is referred to as rising upper end detection, and (d) is referred to as falling upper end detection.

  FIG. 13 shows the trajectory of time and lighting frequency when tracing the lower end of the stable lighting frequency band according to FIG. 11, and FIG. 14 shows the trajectory of time and lighting frequency when tracing the upper end of the stable lighting frequency band according to FIG. Shown in

15 and 16 show the trajectories when the operation switching point is too early and too late in the control for tracing the lower end of the stable lighting frequency band.
As shown in the figure, even when the operation switching point is different from the point where the stable lighting frequency band actually shifts from rising to falling, the stable lighting frequency can be captured again without greatly deviating from the stable lighting frequency band. The same applies when control is performed to trace the upper end of the stable lighting frequency band.
As described above, by controlling to trace the upper end or lower end of the stable lighting frequency band, the stable lighting frequency can be acquired more reliably.

Embodiment 4 FIG.
In the methods shown in the first to third embodiments, the locus of the lamp state parameter and the lighting frequency follows the upper end or the lower end of the stable lighting frequency band while going back and forth between the stable state and the unstable state. The value is a mixture of a stable state and an unstable state. Therefore, in order to actually perform stable lighting, it is necessary to update the stable lighting frequency information in the normal lighting mode.

In the fourth embodiment, all lamp state parameters and lighting frequency trajectories are not stored as stable lighting frequency information, but are stored only in a stable state.
FIG. 17 shows a flowchart for explaining the control method of the initial lighting mode in the fourth embodiment. In this control method, at each predetermined control cycle (Δt), the stability determination signal is used to identify the stable state or the unstable state (S21-S24). If it is stable, changing the lighting frequency by a predetermined value (Δf) (S25-S27) is the same as in the first embodiment, but before the lamp state parameter and the lighting frequency are stored, the stability determination signal is set again. Referring to (S28), the relationship between the lamp state parameter and the lighting frequency is stored as stable lighting frequency information only when the state is stable (S29).

The storage points of the stable lighting frequency in this case are indicated by circles in FIG.
Since the relationship between the lamp parameter and the stable lighting frequency is obtained as a skipped value, in the normal lighting mode, as shown in FIG.

  As described above, according to the method of the fourth embodiment, since only the points actually in the stable lighting frequency band are stored as stable lighting frequency information, stable lighting is performed from the first lighting in the normal lighting mode. Will come to be.

  The flowchart of FIG. 17 and the trajectory of the lamp state parameter and the lighting frequency shown in FIG. 18 correspond to FIG. 6 and FIG. 7 described in the first embodiment, respectively, but in a stable state. Only the method of storing the relationship between the lamp state parameter and the lighting frequency as the stable lighting frequency information can be similarly applied to the method described in the second or third embodiment.

Embodiment 5 FIG.
In the fifth embodiment, trial lighting in the initialization mode is performed by lighting twice. That is, in the first trial lighting, lighting is performed according to the lower limit of the stable lighting frequency, and in the second trial lighting, lighting is performed according to the upper limit of the stable lighting frequency. Then, in each lamp state parameter, a stable lighting is obtained by arithmetically averaging the lower limit value of the stable lighting frequency obtained by the first trial lighting and the upper limit value of the stable lighting frequency obtained by the second trial lighting. Store as frequency information.
Then, as shown in FIG. 20, the vicinity of the center of the stable lighting frequency band can be obtained as the lighting frequency, and more stable lighting can be performed in the normal lighting mode.

  In the above description, lighting is performed in accordance with the lower limit of the stable lighting frequency in the first test lighting, and lighting is performed in accordance with the upper limit of the stable lighting frequency in the second test lighting. Lighting may be performed along the upper limit of the stable lighting frequency, and lighting may be performed along the lower limit of the stable lighting frequency in the second trial lighting.

  In the above description, the arithmetic average of the lower limit value of the stable lighting frequency obtained in the first trial lighting and the upper limit value of the stable lighting frequency obtained in the second lighting is used as the lighting frequency. It is not necessary to be an arithmetic average, and any value that is larger than the lower limit value and smaller than the upper limit value can be used as the lighting frequency. For example, the upper limit value and the lower limit value may be weighted and averaged, or the average value of the drive cycles (average value of the reciprocal frequency) may be used.

It is a block diagram which shows the lighting device of the high-intensity discharge lamp by Embodiment 1 of this invention. It is a circuit diagram which shows the drive circuit in Embodiment 1 of this invention. It is the figure which showed typically the relationship between the elapsed time from the lighting start of a high-intensity discharge lamp, and the stable lighting frequency. It is a flowchart which shows the operation | movement procedure in normal lighting mode in Embodiment 1 of this invention. It is a figure which shows typically the locus | trajectory of the relationship between the elapsed time from the lighting start in Embodiment 1 of this invention, and a lighting frequency. It is a flowchart which shows the operation | movement procedure of the initialization mode in Embodiment 1 of this invention. It is a figure which shows typically the locus | trajectory of the relationship between the elapsed time from the lighting start when using an actual control procedure, and the lighting frequency in the initialization mode of Embodiment 1 of this invention. It is a figure which shows the locus | trajectory of the relationship between the elapsed time from the lighting start in Embodiment 2 of this invention, and a lighting frequency. It is a figure which shows how to make the lighting frequency change with respect to the time domain from the lighting start in Embodiment 2 of this invention. In Embodiment 1 of this invention, it is a figure which shows the locus | trajectory of the relationship between the elapsed time from a lighting start, and a lighting frequency for showing the problem when operation | movement switching is too early. In Embodiment 3 of this invention, it is a figure which shows how to make the lighting frequency change with respect to the operation switching time in the case of tracing the lower end of the stable lighting frequency band. In Embodiment 3 of this invention, it is a figure which shows how to make the lighting frequency change with respect to the operation switching time in the case of tracing the upper end of the stable lighting frequency band. It is a figure which shows the locus | trajectory of the relationship between the elapsed time from the lighting start in Embodiment 3 of this invention, and a lighting frequency. It is a figure which shows the locus | trajectory of the relationship between the elapsed time from the lighting start in Embodiment 3 of this invention, and a lighting frequency. In Embodiment 3 of this invention, it is a figure which shows the locus | trajectory of the relationship between the elapsed time from the lighting start and lighting frequency when operation | movement switching is too early. In Embodiment 3 of this invention, it is a figure which shows the locus | trajectory of the relationship between the elapsed time from the lighting start and lighting frequency when operation | movement switching is too late. It is a flowchart which shows the operation | movement procedure of the initialization mode in Embodiment 4 of this invention. It is a figure which shows the relationship between the elapsed time from starting and the stable lighting frequency memorize | stored as stable lighting frequency information in Embodiment 4 of this invention. It is a figure which shows the locus | trajectory of the lighting frequency in the normal lighting mode of Embodiment 4 of this invention. It is a figure which shows the relationship between the elapsed time from starting and the stable lighting frequency memorize | stored as stable lighting frequency information in Embodiment 5 of this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 High-intensity discharge lamp, 2 Power supply circuit, 3 Drive circuit, 4 Lamp current detection circuit, 5 Lamp voltage detection circuit, 6 Control circuit, 61 Power control block, 62 Stable state determination block, 63 Drive frequency control block, 64 Stable lighting Frequency storage block, 65 stable lighting frequency acquisition block, 66 control signal generation block

Claims (9)

In a lighting device for a high-intensity discharge lamp whose stable lighting frequency varies depending on the internal state,
Stable state determination means for determining whether or not the high-intensity discharge lamp is lit stably and outputting a stability determination signal indicating whether it is a stable state or an unstable state;
Based on the stability determination signal, trial lighting of the high-intensity discharge lamp is performed while changing the lighting frequency so as to avoid an unstable state, and the lamp state parameter indicating the lighting state at that time and the locus of the lighting frequency are stored. Stable lighting frequency acquisition means for acquiring stable lighting frequency information,
A lighting device for a high-intensity discharge lamp, comprising: driving means for driving the high-intensity discharge lamp at a lighting frequency determined based on the stable lighting frequency information.   2. The lighting device for a high-intensity discharge lamp according to claim 1, wherein the lamp state parameter is one of an elapsed time from the start of lighting, a lamp voltage, and a lamp impedance.   3. The high luminance according to claim 1, wherein, in the trial lighting, a rate of change of stable lighting frequency information with respect to time is changed in accordance with a plurality of time regions divided by an elapsed time from the start of lighting. Discharge lamp lighting device.   4. The stable lighting frequency acquisition unit according to claim 1, wherein only a point in a stable state is stored as stable lighting frequency information among the lamp state parameters and the locus of the lighting frequency. The high-intensity discharge lamp lighting device described. In the lighting method of the high-intensity discharge lamp whose stable lighting frequency band changes depending on the internal state,
Based on a stability determination signal indicating whether or not the high-intensity discharge lamp is stably lit, trial lighting is performed while changing the lighting frequency for the high-intensity discharge lamp so as to avoid an unstable state, and the high-intensity discharge is performed. An initialization mode for acquiring, for each high-intensity discharge lamp, stable lighting frequency information indicating the relationship between the lamp state parameter indicating the lighting state of the entire period from the start of lamp lighting to the steady state and the stable lighting frequency, and
And a normal lighting mode in which normal lighting is performed at a stable lighting frequency based on the stable lighting frequency information. 6. The stable lighting frequency information is obtained by starting lighting with a predetermined lighting frequency as an initial value and changing the lighting frequency by any one of the following methods (a) to (d). Lighting method for high-intensity discharge lamps.
I. Ascending upper end detection: When in an unstable state, the lighting frequency is maintained as it is or is gradually increased, and when in a stable state, the lighting frequency is increased more rapidly than in an unstable state.
B. Lower end detection: When in a stable state, the lighting frequency is kept as it is or slowly lowered, and when in an unstable state, the lighting frequency is lowered more steeply than in a stable state.
C. Rising and falling edge detection: When in a stable state, the lighting frequency is kept as it is or gradually increased, and in an unstable state, the lighting frequency is increased more rapidly than in a stable state.
D. Lower end detection: When in an unstable state, the lighting frequency is kept as it is or slowly lowered, and when in a stable state, the lighting frequency is lowered more steeply than in an unstable state.   After the start of lighting, the above-mentioned a (upward upper end detection) is performed until the lamp state parameter reaches a predetermined value, and the lower (upward upper end detection) is performed after the lamp state parameter reaches a predetermined value. The method for lighting a high-intensity discharge lamp according to claim 6.   After starting the lighting, the above-mentioned c (rising lower end detection) is performed until the lamp state parameter reaches a predetermined value, and the d (falling lower end detection) is performed after the lamp state parameter reaches a predetermined value. The method for lighting a high-intensity discharge lamp according to claim 6.   The above-mentioned initialization mode test lighting is performed twice, the first test lighting performs lighting along the lower limit or upper limit of the stable lighting frequency, and the second trial lighting satisfies the upper limit or lower limit of the stable lighting frequency. A value obtained by arithmetically averaging the lower limit value and the upper limit value of the stable lighting frequency obtained by each trial lighting in each lamp state parameter after lighting is stored as stable lighting frequency information. The lighting method of the high-intensity discharge lamp according to any one of 5 to 8.

Images