JPH03167796A - Discharge lamp lighting device - Google Patents

Abstract

PURPOSE:To prevent increase in the circuit loss and generation of leading phase mode resulting from accumulated energy in a resonant circuit by setting the transfer speed from No.2 to No.1 frequency good lower than the transfer speed from No.1 to No.2 frequency. CONSTITUTION:A frequency control means which changes over the operating frequencies of switching elements Q1, Q2 from No.1 frequency f1 higher than the resonance frequency fc of a resonant circuit to No.2 frequency f2 which is higher than the resonance frequency fc but lower than No.1 frequency f1, and vice versa alternately, when a lighting judgement device B decides no lighting a condition of discharge lamp R1. Then the transfer speed from No.2 frequency f2 to No.1 frequency f3 is set sufficiently lower than the transfer speed from No.1 frequency f1 to No.2 frequency f2. This process lessens the circuit loss and prevents generation of the leading phase mode resulting from free oscillation of the resonance circuit.

Description

Detailed Description of the Invention [Field of Industrial Application] The present invention relates to a discharge lamp lighting device using a resonant inverter circuit, and for example, a high intensity discharge lamp (HID).
) is suitable for lighting at high frequency.

[Prior Art] Figure 6 is a circuit diagram of a conventional example. The circuit structure will be explained below. A series circuit of switching elements Q, Q2 is connected in parallel to the DC power supply V,. Each switching element Q, Q2 is composed of a power MOSFET and has a parasitic anti-parallel diode. A resonance capacitor C is connected to both ends of the switching element Q2 via a current limiting inductor L and a DC cut capacitor C2. The resonance capacitor C1 has
The discharge lamp R. are connected in parallel. Each switching element Q 1, Q
2 are alternately driven on and off by drive circuits DRI and DR2. As a result, inductor L and capacitor C
An approximately sinusoidal alternating current flows from I to a certain resonant circuit,
The AC voltage generated across the capacitor C1 is the discharge lamp RL.
is applied to. The capacitance of the DC cut capacitor C2 is set to be sufficiently larger than the capacitance of the resonance capacitor Ct, and does not contribute to resonance. In a steady state, when the on periods of the switching elements Q, , Q2 are equal, a voltage of 1/2 of the DC power supply V1 is charged to the DC cut capacitor C2.

Discharge lamp R. The primary winding of a current transformer CT is connected in series with . One end of the secondary winding of the current transformer CT is grounded, and the other end is connected to one end of a capacitor C via a rectifying diode D1. The other end of capacitor C is grounded. The voltage obtained across capacitor C is applied to the non-inverting input terminal of converter CMP. A reference voltage Vr is applied to the inverting input terminal of the converter CMP. The output VcP of the converter CMP is PNP} transistor Q. is supplied to the base. This PNP}-transistor Qo is interposed between the oscillator T1 for high frequency oscillation and the oscillator T2 for low frequency oscillation, and opens and closes a control signal from the latter to the former.

The oscillator T1 for high frequency oscillation consists of a general-purpose timer C (for example, NE555 manufactured by Signetics). This timer IC is externally connected with resistors R + , R 2 and capacitor C4 to prevent unstable multi-vibrators, and the control voltage V applied to its frequency control terminal (bin 5) increases. Then, the frequency of the oscillation output V, decreases. The oscillation output V of the oscillator T1 is used as one input of the first and second AND circuits AND1 and AND2, and is also used as the clock input C of the D flip-flop FF. The output Q of the D flip-flop FF is connected to the data input D. Therefore, the output Q of the D flip-flop FF, ? :A is obtained by dividing the oscillation output V of the oscillator T1 into 1/2 frequency. The outputs Q and Q of this D flip-flop FF are used as the other inputs of the first and second AND circuits ANDI and AND2, respectively. AND circuit ANDI, AND2
Each output VDVD2 of the drive circuit DR. 1,D
Switching elements Q, . It is supplied to the control electrode of Q2. Next, the oscillator T2 for low frequency oscillation is composed of the same general-purpose timer IC (NE555 manufactured by Sidanetics) as the oscillator T1 for high frequency oscillation, and is connected to resistors R, , R to form an astable multivibrator. and capacitor C
, is attached externally. The frequency control terminal (pin 5) of this oscillator T2 is grounded via a decoupling capacitor C6. Therefore, its oscillation frequency is fixed and set sufficiently lower than the oscillation frequency of the oscillator T1. The output terminal (bin 3) of the oscillator T2 is connected to the input terminal of the inverter INV.
The output terminal of the inverting circuit INV is connected to the resistor R via the resistor R.
,,R. connected to one end of each.

The other end of the resistor R5 is connected to the control power supply voltage Vcc, and the other end of the resistor R6 is grounded. The connection point of resistors R5R6 is the above-mentioned transistor Q. connected to the emitter of
Transistor Q. The collector of is connected to the frequency control terminal (bin 5) of the oscillator T1. Figure 7 shows the resonance characteristics of the above circuit. The horizontal axis f is the operating frequency of the inverter device, and the vertical axis VRL is This is the voltage across the discharge lamp RL. In a no-load state where the discharge lamp RL is not lit, the impedance of the discharge lamp RL is substantially infinite. At this time, near the resonance frequency f. The voltage VRL of RL becomes very high.Therefore, in order to utilize this characteristic for starting the discharge lamp RL, in a no-load state, the operating frequency f of the inverter device resonates with a frequency f far from the resonance frequency 10. frequency f0
It is alternately switched to a frequency f2 close to . and,
When operated at a frequency of 12 close to the resonant frequency f0, the discharge lamp R. A high voltage for starting is applied to the On the other hand, when the discharge lamp RL lights up, the discharge lamp R. Since the impedance of the resonant curve decreases, the slope of the resonant curve becomes gentler, and the resonant frequency decreases. Therefore, in this case, the inverter device is operated at the frequency f3 at the time of lighting. In either case, since the resonant circuit is excited at a frequency higher than its resonant frequency, the resonant current flowing through the inductor is delayed in phase with respect to the excitation voltage generated across switching element Q2. This is called the slow mode.
In this slow phase mode, switching element Q is turned off and switching element Q2 is turned on at the timing when forward current is flowing through switching element Q1, and the resonant current flowing in the forward direction of switching element Q1 is transferred to switching element Q2. It flows through the reverse direction diode, and then, when the polarity of the resonant current is reversed, it flows in the forward direction of the switching element Q2, which is already turned on. Switching element Q
The same applies to the case where Q1 is on and switching element Q2 is off. Therefore, in slow phase mode,
Switching element Q. (or Q2) is turned on, a reverse current is flowing through the switching element Q, (or Q2), and the switching elements Q1, Q2 are turned on.
There will be no inconvenience such as two being turned on at the same time.

If the inverter device is operated at the lighting frequency f3 in a no-load state, the resonant circuit will be excited at a frequency lower than its resonant frequency f0, so the resonant current flowing through the inductor will flow across the switching element Q2. It has a leading phase with respect to the excitation voltage generated in . This is called the phase advance mode. In this phase advance mode, switching element Q1
When the reverse current is flowing through the switching element Q. is off, switching element Q2 is turned on, and the resonant current flowing through the reverse direction diode of switching element Q1 flows in the forward direction of switching element Q2. The same applies when switching element Q1 is turned on and switching element Q2 is turned off. In this way, in the phase advance mode, at the timing when switching element Q+ (or Q2) is turned on, the other switching element Q. (or Q
Since a reverse current is flowing through the switching element Q l, ), the switching element Q l,
This causes the inconvenience that Q2 is turned on at the same time. Therefore, in the circuit shown in FIG. 6, when the discharge lamp RL is in a no-load state, the inverter device is operated at the frequency f1 or f.
2, and after determining that the discharge lamp RL is in the lighting state, the inverter device is operated at a frequency f, and by always operating in the slow phase mode, the switching elements Q, . This prevents Q2 from turning on at the same time. The operation will be explained in detail below. FIG. 8 is an operating waveform diagram of the above circuit, showing the voltage V of the capacitor C, which determines the oscillation frequency of the oscillator T1. , and the oscillation output V of the oscillator T1, and the drive circuit IiI8DR1,D
AND circuits ANDI and AND each input to R2
2 shows the relationship between the outputs VDI and VD2 of the converter T2, the voltage Vx controlled by the oscillator T2, and the output VcP of the converter CMP. Discharge lamp R. When is in the lighting state, a current IRL flows through the discharge lamp R, so the voltage of the capacitor C increases, and the voltage becomes equal to the reference voltage Vr.
By exceeding V, the output of the converter CMP. P
becomes “High” level. Therefore, PNP} transistor Q. becomes non-conductive. Therefore, the oscillation frequency of oscillator T1 is determined by resistor R. , R2 and the capacitor C, the frequency at the time of lighting is 13. Further, when the discharge lamp RL is in a no-load state, the voltage of the capacitor C becomes lower than the reference voltage V', and the output VCP of the converter CMP becomes a 'Low' level. Therefore, the PNP transistor Q0 becomes conductive. Therefore, the control voltage V, applied to the frequency control terminal (pin 5) of oscillator T1 becomes equal to the voltage Vx controlled by oscillator T2. ) is at “High” level, the output of the inverter INV is “Log
” level, so the resistor R7 is connected in parallel with the resistor R6, and the voltage Vx becomes low.On the other hand, the oscillator T
When the output terminal 2 (pin 3) is at the "Low" level, the output of the inverting circuit INV is at the "High" level, so the resistor R7 is connected in parallel with the resistor R, and the voltage v× 5 This causes the oscillation frequency of the oscillator T1 to be switched. Figure 9 is an operating waveform diagram of the above circuit in a no-load state.
1 shows the relationship between the oscillation frequency f of No. 1 and the voltage VRL across the discharge lamp RL. When the voltage Vx is low, the oscillator T1
The oscillation frequency becomes a high frequency f1, and is away from the no-load resonance frequency f. At this time, the voltage V across the discharge lamp RL
RL is low. Also, when the voltage Vx is high, the oscillator T
The oscillation frequency of 1 becomes a low frequency f2, and the no-load resonance frequency f. approach. At this time, the discharge lamp R. The voltage across V
RL becomes higher. The oscillation frequency of oscillator T1 is frequency f2
The period set for frequency f, is set shorter than the period set for frequency f. Therefore, for the discharge lamp RL,
A high voltage pulse is applied at regular intervals determined by the oscillator T2. This high voltage pulse starts the discharge lamp RL. However, in this conventional example, in a no-load state, the resonant frequency f. After generating a high voltage pulse at a frequency f2 close to the resonant frequency f. When returning to frequency 11, which is far from , the phase advance mode sometimes occurred. This is the resonant frequency f. When operating at a frequency f2 close to , generating a high voltage pulse, a large amount of energy is accumulated in the resonant circuit including the inductor L and capacitor C1, and after returning to the frequency f1, the above energy is released by free vibration. This is thought to be due to the Operation in such a phase advance mode continues until the energy of the resonant circuit is attenuated, but since the discharge lamp RL has a high impedance in a no-load state, the energy of the resonant circuit is difficult to attenuate, and the phase advance is delayed. The phase mode is likely to persist. Therefore, the switching element Q. ．． Q. Simultaneously turning on caused an overcurrent to flow, which could lead to deterioration or destruction of the device. Figure 10 is a circuit diagram of another conventional example. In this circuit, the oscillation frequency of the high frequency oscillator A that supplies drive signals to the drive circuits DRI and DR2 is made variable by the outputs of the lighting discriminator B and the low frequency oscillator C. That is, when the lighting discriminator B determines that the discharge lamp RL is in the lighting state, the oscillation frequency of the high-frequency oscillator A is set to the lighting frequency f, and the lighting discriminator B determines that the discharge lamp RL is in the no-load state. When it is determined that the device is in the non-lighting state, the oscillation frequency of the high frequency oscillator A is alternately switched to a frequency f1 far from the resonance frequency f0 and a frequency f2 close to the resonance frequency r0. However, in order to smoothly switch between the frequencies f and f2, a capacitor Cp is connected to the output of the low frequency oscillator C. In this way, by smoothly switching between frequency 11 and frequency f2 in the no-load state, the energy of the resonant circuit is gradually increased,
Thereafter, it can be gradually attenuated, thereby preventing the occurrence of a phase advance mode due to a sudden change in frequency. When the discharge lamp RL is turned on, the output of the low frequency oscillator C is short-circuited by the lighting discriminator B, and the high frequency oscillator A oscillates at a frequency of 13 at the time of lighting. The specific circuit configuration will be explained below. First, the low frequency oscillation hc has basically the same configuration as the oscillator T2 used in Conventional Example 1, only the inverting circuit INV is omitted. Next, the lighting discriminator B will be explained. A low resistance Rs for current detection is connected in series to the discharge lamp RL. The voltage generated across this low resistance Rs is applied to the non-inverting input terminal of the comparator CMP. A reference voltage V' is applied to the inverting input terminal of the comparator CMP. The output voltage of comparator CMP is charged to capacitor C via resistor R0 and diode D1. The voltage Vd obtained across the capacitor C is the voltage Vd obtained across the resistor R8.
is supplied to the base of transistor Q6 via.

Transistor Q6 is connected in parallel to both ends of capacitor Cp. As described above, the lighting discriminator B is elliptical. Next, high frequency oscillator A will be explained. Capacitor Cp
The voltage Vp obtained at
is supplied to the base.

Here, the operational amplifiers A to IP are amplifiers with high input impedance, low output impedance, and a gain of 1, and act as buffers to prevent the voltage Vp of the capacitor Cp from being influenced by the high-frequency oscillator A. do. The emitter of transistor Q5 is resistor R,. The collector is connected to a resistor R. The resistor R is connected to one end of the resistor R, and the other end of the resistor R is grounded. Resistance R. The one end of the transistor Q3 is connected to the base and collector of the PNP transistor Q3, and is also connected to the base of the PNP transistor Q4. PNP} transistors Q3, Q, have their collectors connected to the control fl4t source voltage Vcc.

PNP} The collector of transistor Q4 is connected to capacitor CF.
and one end of each resistor R+2, and is also connected to the input terminal of the Schmitt circuit ST. capacitor CF
The other end of the resistor Rl2 is grounded, and the other end of the resistor Rl2 is grounded via the transistor Q. The base of transistor Q7 is connected to the output terminal of Schmitt circuit ST via resistor R1. In the Schmitt circuit ST, when the voltage VCF obtained across the capacitor CF exceeds a first level, the output VST becomes a "High" level, and when it falls below a second level lower than the first level, the output VST becomes a "High" level.
becomes the 'L01' level. The output vs'r of this Schmitt circuit ST is used as one input of the AND circuits AND1 and AND2, and is also used as the clock input C of the D flip-flop FF. The negative output Q of the D flip-flop FF is connected to the data output D. As a result, the output Q and the negative output Q of the D flip-flop FF
In this case, a signal obtained by dividing the clock C by half the frequency is obtained.

Each output Q, Q of the D flip-flop FF is used as the other input of the AND circuits ANDI, AND2. The respective outputs of the AND circuits AND1 and AND2 are supplied to the control electrodes of the switching elements Q, Q2 via drive circuits DRI and DR2, respectively.

FIG. 11 is an operating waveform diagram of the above circuit, and shows the lighting discriminator B.
The voltage Vd of the capacitor C, and the low frequency oscillator C
The voltage Vp of the capacitor Cp connected to the output of the discharge lamp R., the current Ie flowing through the transistor Q, the oscillation frequency f of the high frequency oscillator A, and the discharge lamp R. It shows the relationship between the voltage VRL across the . The operation of the above circuit will be explained below.

First, in a no-load state, the current IRL in the discharge lamp R[
does not flow, no voltage is generated across the low resistance Rs for current detection, and the output of the comparator CMP is “Low”.
" level, and the voltage Vd of the capacitor C is low. Therefore, the transistor Q6 is turned off, and the voltage Vp of the capacitor Cp increases and decreases at a constant period according to the output of the low frequency oscillator C.

Oscillator T consisting of a timer IC in low frequency oscillator C
When the output terminal 2 (bin 3) is at the "High" level, the resistor R is connected in parallel to the resistor R, and in a steady state, the voltage Vp of the capacitor cp is higher than the control power supply voltage Vcc. A high voltage Vpl is obtained by dividing the voltage by the parallel resistor (RS//R, ) and R6. At this time, the oscillation frequency of high frequency oscillator A is f=f. becomes. Next, the oscillator T
When the output terminal 2 (pin 3) goes to "Low" level, the resistor R is connected in parallel to the resistor R6, and the voltage Vp of the capacitor Cp connects the control g4 power supply voltage Vcc in parallel with the resistor R. The low voltage Vl)2 divided by the resistor (Ra//R, ) is set as a target value and decreases at a predetermined time constant. V
When l)1 and l)2, the oscillation frequency of high frequency oscillator A is r= r. becomes. Output terminal of oscillator T2 (bin 3)
The period in which the capacitor c is at the “Low” level is set short and returns to the “High” level immediately.
The voltage Vp of p is again set to the above voltage Vp+ as the target value,
It increases with a predetermined time constant. At this time, the oscillation frequency r of the high frequency oscillator A is the frequency f. The frequency gradually decreases from 1 to 2 to the frequency f2, and then slowly increases to the frequency f.

Therefore, the voltage VRL across the discharge lamp RL increases rapidly and then gradually decreases.

The oscillation operation of the high frequency oscillator A is performed by a Schmitt circuit ST. A capacitor CF connected to the input terminal of the Schmitt circuit ST is charged by the current flowing through the transistor Q.

When the charging pressure VCF of the capacitor CF exceeds the first level, the output V5 of the Schmitt circuit ST becomes "Hgh" level. At this time, a base current flows to the transistor Q7 via the resistor Rl3, and the transistor Q7 is turned on. Therefore, a resistor R12 is connected in parallel to the capacitor CF, and the electric charge of the capacitor CF is discharged through the resistor R+2, so that its charging voltage VCF decreases. A second level in which the charging voltage VCF of the capacitor CF is lower than the first level.
below the level of VST, the output of Schmitt circuit ST
becomes “Low” level. At this time, the transistor Q7 is turned off, and the discharge of the capacitor CF is stopped.
Capacitor CF is charged again by the current flowing through transistor Q4. Thereafter, the same operation is repeated, and an oscillation output consisting of a rectangular wave voltage is obtained as the output VST of the Schmitt circuit ST. This oscillation output is frequency-divided by the D flip-flop 1FF, and the AND circuit AND1,A
Dead time is given by ND2, and the drive circuit DR
I, the switching element Q. via DR2. , becomes the drive signal for Q2. Since the oscillation frequency of the Schmitt circuit ST is determined by the charging/discharging rate of the capacitor CF, the oscillation frequency can be changed by controlling the current flowing through the transistor Q4. Transistor Q. ,Q.
Is hfe almost equal and the current mirror circuit is used as a tank? Therefore, the current flowing through transistor Q is the same as the current Ic flowing through transistor Q.

Since this transistor Q has its base and collector short-circuited and is used as a simple PN junction diode, its collector-emitter voltage VCE3 is approximately constant (
0.6 to 0.7V). Therefore, transistor Q and resistor R. The voltage VRI1 at the connection point is a constant value (Vc
c Vc2,), and the resistance R. The current IRI flowing through
+ is also a constant value VR■/R. becomes. The current Ic flowing through the transistor Q is the sum of the current I Rl1 flowing through this resistor R and the current If flowing through the transistor Q5, and the current If is variable according to the voltage Vp applied from the outside. There is. When the t-pressure Vp rises, the output voltage Va of the operational amplifier AMP rises, and the transistor Q
Since the emitter potential Vf of transistor Q5 increases, the emitter potential Vf of transistor Q
5 through resistor R1. Current If (=Vf/R1
o) increases. Conversely, when voltage Vp decreases, current If decreases. Therefore, as shown in the operating waveform diagram of FIG. 11, when the voltage Vp decreases, the current I c < - I
Rll+I f) decreases and the charging speed of the capacitor CF becomes slower, so the oscillation frequency f becomes lower. Conversely, when the voltage Vρ increases, the current Ic increases and the charging speed of the capacitor CF becomes faster, so that the oscillation frequency f becomes higher.

Next, when the discharge lamp RL enters the lighting state, the current IRL flows, so the output of the comparator CMP changes at high frequency.
The state is such that the "high" level and the "low" level are repeated, and the voltage Vd of the capacitor C3 increases.

Therefore, transistor Q6 is turned on, capacitor Cp is short-circuited, and its voltage Vp becomes approximately zero. At this time, the oscillation frequency f of the high frequency oscillator A becomes the frequency r3 at the time of lighting, and is fixed as it is.

In this conventional example, in a no-load state, the voltage Vp for switching and controlling the oscillation frequency r of the high-frequency oscillator A is slowly changed by the capacitor Cp, so that the resonant frequency f. From the frequency F1 far from the resonance frequency r. The energy in the resonant circuit is gradually accumulated as the frequency f2 is gradually approached, and then the energy in the resonant circuit is gradually released as the frequency is gradually moved away from the frequency f2 to the frequency 11. Therefore, the phase advance mode is never established.

However, in this conventional example, the voltage VR of the discharge lamp RL
Since L increases slowly and then decreases slowly, the period during which a large current flows in the resonant circuit becomes longer, increasing circuit loss, especially switching loss, and as a result, the switching element Q ,,Q. deterioration cannot be prevented. Furthermore, since the high voltage is applied to the resonant circuit for a long period of time, it is necessary to use expensive components with high voltage resistance for the circuit elements, which increases the cost of the device. [Problems to be Solved by the Invention] As described above, when lighting a discharge lamp using a resonant inverter device, a high voltage can be generated by bringing the operating frequency close to the resonant frequency of the resonant circuit. Therefore, by utilizing this property, as in Conventional Example 1, in the no-load state of the discharge lamp, the operating frequency of the inverter device is set to the resonance frequency f. the more distant first frequency f. It has been proposed to apply periodic high voltage pulses to the discharge lamp by repeating at predetermined intervals the operation of increasing the second frequency r2 close to the resonance frequency 10, and to start the discharge lamp with the high voltage pulses. ing. However, the first frequency f
, the energy stored in the resonant circuit is small, and the second
Since the energy stored in the resonant circuit is large at the frequency f2 of , if the operating frequency is suddenly switched, the energy of the resonant circuit may be released by free vibration, resulting in a phase advance mode, and the simultaneous switching of the switching elements. When the switch is turned on, an overcurrent flows, which can lead to element deterioration or destruction. Therefore, it has been proposed to switch the operating frequency slowly as in Conventional Example 2. However, conventional example 2
Although it is possible to prevent the phase advance mode from occurring, the period during which large amounts of energy are stored in the resonant circuit becomes longer, resulting in the problem of increased circuit loss.Also, as the current flowing through the switching element increases, Since the switching voltage also increases, there is a problem in that it is not possible to prevent element deterioration as a result. The present invention has been made in view of the above points, and its purpose is to reduce the operating frequency to resonance in a no-load state in a discharge lamp lighting device that lights a discharge lamp using a resonant inharter device. The object of the present invention is to avoid the occurrence of a phase advance mode caused by the energy stored in a resonant circuit while preventing an increase in circuit loss when applying a high voltage for starting to a discharge lamp close to the frequency.

[A Thousand Steps to Solve the Problem] In order to solve the above problem, the present invention has the following steps:
As shown in the figure, the first and second
The switching element Q. , Q2 series circuit is connected to the DC power supply V
1, and the first and second switching elements Q. ．． Q. A drive circuit DR.
, DR2, including an inductor L, a capacitor C1, and a discharge lamp RL, first and second switching elements Q. ，
In the discharge lamp lighting device including a resonant circuit excited by the voltage obtained at the connection point of the discharge lamp R. A lighting discriminator B that determines whether or not the lamp is in a lighting state, and a discharge lamp RL.
When it is determined that the switching elements Q, . The operating frequency of Q2 is set to the resonant frequency f0 of the resonant circuit.
a first frequency higher than f. and a second frequency f2 that is higher than the resonance frequency f0 and lower than the first frequency r1, and the transition speed from the second frequency f2 to the first frequency f1 is as follows: It is characterized in that it is set sufficiently slower than the transition speed from the first frequency f to the second frequency f2.

[Function] According to the present invention, in the discharge lamp lighting device using a resonance type inverter device, the discharge lamp R. When it is determined that is not in the lighting state, the operating frequency r of the switching elements Q, , Q2 is set to the first frequency f, which is higher than the resonant frequency f0 of the resonant circuit, and the resonant frequency f. , and a second frequency 12, which is lower than R. It provides high voltage for starting. At this time, when changing from the first frequency 11 to the second frequency r2, the frequency was changed rapidly.
Compared to the case where the frequency is changed slowly as in Conventional Example 2, the circuit loss is reduced. At this time, since the resonant circuit is in the stage of increasing stored energy, a phase advance mode due to free vibration does not occur.

Next, when returning from the second frequency 12 to the first frequency r, the resonant circuit is in the stage of reducing the stored energy, so a phase advance mode may occur due to free vibration of the resonant circuit, but the present invention In this case, as in Conventional Example 2, since the frequency is changed slowly, the phase advance mode caused by free vibration of the resonant circuit can be prevented.

In other words, when the stored energy of the resonant circuit increases, the resonant circuit enters a forced oscillation state that is easily dominated by external excitation voltage, but when the stored energy of the resonant circuit decreases, the resonant circuit enters a forced oscillation state that is easily dominated by external excitation voltage. It becomes a free vibration state that is not easily controlled by voltage. In the free vibration state, even if the excitation frequency is higher than the resonant frequency, the phase advance mode may occur, but in the forced vibration state, if the excitation frequency is higher than the resonance frequency, the phase advance mode is unlikely to occur. The present invention focuses on the properties of such a resonant circuit, and changes the frequency slowly only in the free vibration state where the phase advance mode is likely to occur, and the forced vibration where the phase advance mode is unlikely to occur. By changing the frequency rapidly in the state, it is possible to reduce circuit loss compared to the conventional example 2 in which the frequency is always changed slowly. [Embodiment 1] FIG. 1 is a circuit diagram of a first embodiment of the present invention.

The circuit configuration will be explained below. DC power supply ■1 and switching element Q. , Q2, the ellipse of the main circuit including the inductor, capacitor C 1C 2 and discharge lamp RL is:
Since it is similar to the conventional example shown in FIG. 6 or 10,
Duplicate explanations will be omitted. In addition, the lighting discriminator B is the first
Output VDET corresponding to voltage Vd of the conventional example shown in Figure 0
is configured to occur.

In this embodiment, there are three oscillators OSCI to OSC3.
It is equipped with The first oscillator OSC1 is the second oscillator O
This is an oscillator for periodically changing the control voltage Vc applied to SC2. The second oscillator ○SC2 has an oscillation frequency re of f1 to f1 according to the output voltage of the first oscillator OSCI.
This is a voltage controlled oscillator that varies within the f2 range. Third
The oscillator OSC3 is an oscillator that oscillates at a frequency f during lighting. The oscillation output of oscillator OSC3 is AND circuit A
One input of the oscillator ND3 is used, and the oscillation output of the oscillator OSC2 is used as one input of the AND circuit AND4. The output V DET of the lighting discriminator B is supplied to the other input of the AND circuit AND3, and the AND circuit AN
A signal obtained by inverting the output V DET of the lighting determination B by the inverting circuit INV2 is input to the other input of D4. The output of the AND circuits AND3 and AND4 is the OR circuit O
It is input to the frequency divider DfV via R. Frequency divider DI
The first and second divided outputs of V are respectively output from the drive circuit DR.
It is supplied to the control electrodes of switching elements Q, Q2 via I and DR2. Figure 2 is an operating waveform diagram of this embodiment, and shows the oscillator OSC.
1, the oscillation frequency fc of the oscillator OSC2, the oscillation frequency f3 of the oscillator OSC3, the output VDET of the lighting discriminator B, the output frequency f of the OR circuit OR, and the voltage across the discharge lamp RL. It shows the relationship between VRL. When the power is turned on, the output V of lighting discriminator B
Since DET is at “Low” level, the AND circuit AN
D3 is in a state where signal passage is prohibited, and AND circuit AND4 is in a state where signal passage is possible. Therefore, the output frequency f of the OR circuit OR is determined by the oscillation frequency fc of the oscillator OSC2.

This oscillation frequency fc decreases to a frequency r2 close to the resonance point at regular intervals determined by the oscillator OSCI, and each time, the voltage VRL across the discharge lamp RL increases. by this,
A high voltage pulse for starting is periodically applied to the discharge lamp RL. When the discharge lamp RL is lit by this high-voltage pulse, the output VDET of the lighting discriminator B becomes ``High'' level, so the add circuit AND3 becomes in a state where signals can pass, and the ant circuit AND4 becomes in a state where signals are prohibited from passing. Become. Therefore, the output frequency f of the OR circuit OR is determined by the oscillation frequency r3 of the oscillator SC3.

In this embodiment, the control voltage Vc output from the oscillator OSCI is a sawtooth wave as shown in FIG.
The oscillation frequency re of the oscillator OSC2 changes rapidly from frequency r1 to frequency 12, and slowly changes from frequency f2 to frequency f. Therefore, compared to the conventional example 2 in which the frequency is changed slowly from the frequency f to the frequency r2, the discharge lamp R
The period during which the voltage VRL across L is at a high voltage is shortened, and circuit loss is reduced.

Further, since the change from frequency f2 to frequency r is performed slowly in accordance with the attenuation rate of the energy stored in the resonant circuit, a phase advance mode due to free vibration of the resonant circuit does not occur.

[Example 2] FIG. 3 is a circuit diagram of a second embodiment of the present invention.

In this embodiment, an operational amplifier A is added to the conventional example shown in FIG.
MP2 and resistance R. and diode D2 are added. The operational amplifier A M P 2 is an impedance conversion amplifier having a high input impedance, a low output impedance, and a gain of 1. When the street light RL is not lit, the output impedance of the lighting discriminator B is high, and the voltage Vp of the capacitor Cp is controlled by the low frequency oscillator C. When the output voltage of the low frequency oscillator C becomes low, the capacitor cp is rapidly discharged through the diode D2, so that its voltage Vp rapidly decreases. Therefore, the oscillation frequency f of the high-frequency oscillator A rapidly decreases from frequency f to frequency f2. Also, when the output voltage of the low frequency oscillator C becomes high, the capacitor Cp becomes the resistor R. The voltage Vp rises slowly as the voltage Vp is slowly charged. Therefore, the oscillation frequency r of the high frequency oscillator A slowly changes from frequency f2 to frequency r1. Other operations are similar to the conventional example shown in FIG. In this embodiment, the present invention can be easily implemented by simply adding a few circuit components to the conventional example. [Embodiment 3] FIG. 5 is a circuit diagram of a third embodiment of the present invention.

The circuit configuration will be explained below. DC power supply V; and switching element Q. Q2, inductor and capacitor C. , C2 and discharge lamp R. The main circuit configuration including
Since it is similar to the conventional example shown in FIG. 6 or 10,
Duplicate explanations will be omitted. The lighting discriminator B also uses the discharge lamp R. Any circuit can be used as long as it can determine whether or not the light is on. The output of this lighting discriminator B is
It is used as a control signal for the changeover switch SW. Discharge lamp R.
When it is determined that the discharge lamp RL is not in the lighting state, the changeover switch SW is connected to the terminal a side, and when it is determined that the discharge lamp RL is in the lighting state, the changeover switch SW is connected to the terminal b side. An oscillator OSC is connected to the terminal a side, and a reference voltage Vr is connected to the terminal b side. The oscillator OSC is the same as the oscillator OSC1 described in the first embodiment, and has a sawtooth wave control voltage Vc as shown in FIG.
is generated. This control voltage Vc becomes a voltage that controls the pulse width of the monostable multivibrator MM, and the control voltage V
As c increases, the pulse width becomes wider. The output Q of the monostable multivibrator MM is an AND circuit AND 1 ,
It is used as one input of AND2, and is also used as the single input T that triggers the T flip-flop FF. The output Q and negative output Q of the T flip-flop 1FF are obtained by dividing the trigger force T into 1/2. These outputs are used as one input of AND circuits AND3 and AND4. AND circuit AND3. The outputs of the detectors ZD2 and ZDI are respectively supplied to the other input of AND4. The outputs of the AND circuits AND3 and AND4 are used as the other inputs of the AND circuits AND1 and AND2, respectively, and are also used to trigger the monostable multivibrator MM via the OR circuit OR. Then, the outputs of the AND circuits AND 1 and AND2 are sent to the switching element Q. ，
It is supplied to the control electrode of Q2. In this embodiment, when the phase advance mode is entered, two switching elements Q, . In order to prevent Q2 from turning on at the same time, a simultaneous turning-on prevention circuit is provided which includes a detector ZDI and ZD2 and AND circuits AND3 and AND4. The detector ZDI detects the reverse current flowing through the switching element Q through the current transformer, and the output of this detector ZD1 becomes "'Low" only when the reverse current flows through the switching element Q1. level. Similarly, the detector ZD2 detects the reverse current flowing through the switching element Q2 via the current transformer, and the output of this detector ZD2 becomes "Low" only when the reverse current flows through the switching element Q2. The operation of this embodiment will be explained below. The monostable multivibrator MM is triggered by the rising edge of the output of the OR circuit OR. Then, after a certain period of time has elapsed, the monostable multivibrator MM When the output Q of the T flip-flop FF falls, the output of the T flip-flop FF is inverted.Now, the negative output Q of the T flip-flop 1FF is inverted from the "High" level to the "Lou+ level", and the output Q becomes "Low".
If the level is reversed to “High” level,
The output of the AND circuit AND2 is ``H'' at the falling timing of the output Q of the monostable multivibrator MM.
The level changes from "High" level to "Low" level. Therefore, the switching element Q2 is in a state where it blocks forward current, but at this time, if a reverse current is flowing through the switching element Q1, it is in the slow phase mode and immediately gives an on signal to the switching element Q. However, simultaneous on does not occur. In this case, the output of the detector ZD2 is “High”.
” level, the output of the AND circuit AND 3 becomes the “High” level, and the monostable multivibrator MM is triggered via the OR circuit OR. As a result, the output Q of the monostable multivibrator MM remains high for a certain period of time. becomes a "High" level, during which the output of the AND circuit AND1 becomes a "High" level, and an on signal is given to the switching element Q1. On the other hand, as mentioned above, the output of the AND circuit AND2 is “H”.
If a reverse current is flowing through the switching element Q2 when the level changes from "high" level to the ゜゛L01 level, it is in phase advance mode, and if an on signal is immediately given to the switching element Q1, the simultaneous on phenomenon will not occur. In this case, since the output of the detector ZD2 is at the "Lou+" level, the output of the AND circuit AND3 is at the "Low" level, and the monostable multivibrator MM is not triggered. Thereafter, the reverse current in the switching element Q2 is stopped, and the other switching element Q. When a reverse current flows through the detector ZD2, the output becomes “High”.
level, so at this timing the AND circuit AND
The output of No. 3 becomes ``high'' level, and the monostable multivibrator MM is triggered. by this,
For a certain period of time, the output Q of the monostable multivibrator MM is “
'High' level, during which the AND circuit AND
Is the output of I ``Hii?''h" level, and an on signal is given to the switching element Q1. Therefore, it is possible to prevent the phase advance mode from occurring, and the two switching elements Q1 and Q2 will not be turned on at the same time. The operation has been described for the case where the timing at which the switching element Q1 is turned on is controlled by the AND circuit AND3 and the detector ZD2.
Needless to say, the same holds true for a method in which the timing at which the is turned on is controlled by the AND circuit AND4 and the detector ZDI.

In an inverter device equipped with such a simultaneous-on prevention circuit, the present embodiment is characterized in that the pulse width of the monostable multivibrator MM is made variable by the control voltage (c) applied from the outside. When the power is turned on, the discharge lamp RL is not lit, so the selector switch SW is connected to the terminal a side, and the pulse width of the monostable multivibrator M is controlled by the sawtooth wave control voltage Vc output from the oscillator OSC. is controlled. When the control voltage Vc is high, the pulse width of the monostable multivibrator MM is narrow and the oscillation frequency r becomes the first frequency f. Then, when the control voltage Vc rapidly decreases, the pulse width of the monostable multivibrator MM rapidly widens, and the oscillation frequency f rapidly changes to the second frequency r2. After that, control
When the R voltage Vc rises to M, the pulse width of the monostable multivibrator MM slowly narrows, and the oscillation frequency f becomes the first frequency f. changes slowly. Therefore, the release
Light R. is periodically given high voltage pulses.

This high voltage pulse causes the discharge lamp R. When it starts and lights up, the selector switch SW is connected to the terminal b side,
The pulse width of the monostable multivibrator MM is the reference voltage V
It becomes a constant value determined by r, and the oscillation frequency f becomes the frequency r3 at the time of lighting. In this embodiment, the AND circuit AN
D3, AND4 and detector ZD1. Since it is equipped with a simultaneous ON prevention circuit consisting of ZD2, it is possible to make the frequency change speed when returning from the second frequency f2 to the first frequency f faster than in other embodiments, and the circuit This makes it possible to further reduce losses.

Note that in each of the above embodiments, the switching element Q. ，
Q2 is not limited to power MOSFET,
It may be a bibolar transistor with an anti-parallel diode added, and generally any semiconductor switch element that does not block reverse current can be used.

Further, the circuit configuration of the inverter device is not limited to the modified half-bridge circuit as exemplified in the embodiment, but may also be a half-bridge circuit, a full-bridge circuit, etc., and as long as it is a resonant type inverter device, The present invention can be applied. Here, the full bridge circuit refers to the first and second
A series circuit of switching elements and a series circuit of third and fourth switching elements are connected in parallel to a DC power supply, and a connection point of the first and second switching elements and a series circuit of third and fourth switching elements are connected in parallel to a DC power supply. The third switching element is connected to the second switching element so that voltages of opposite polarity are alternately applied to the LC resonance circuit. The fourth switching element is turned on and off at the same time as the first switching element. Furthermore, a half-bridge circuit is a circuit in which the third and fourth switching elements in a full-bridge circuit are each replaced with a capacitor. [Effects of the Invention] In the present invention, in a discharge lamp lighting device that lights a discharge lamp using a so-called resonance type inverter device, when it is determined that the discharge lamp is not in the lighting state, the operation of the switching element is controlled. When applying a high starting voltage to the discharge lamp by alternately switching the frequency between a first frequency higher than the resonant frequency of the resonant circuit and a second frequency higher than the resonant frequency and lower than the first frequency. In addition, when changing from the first frequency to the second frequency, the frequency is changed rapidly, which has the effect of reducing circuit loss compared to a case where the frequency is changed slowly. When returning from the second frequency to the first frequency, since the frequency is changed slowly, it is possible to prevent a phase advance mode caused by free vibration of the resonant circuit.

[Brief explanation of the drawing]

FIG. 1 is a circuit diagram of a first embodiment of the present invention, FIG. 2 is an operating waveform diagram of the same as above, FIG. 3 is a circuit diagram of a second embodiment of the present invention, and FIG. 4 is an operating waveform diagram of the same as above. FIG. 5 is a circuit diagram of the third embodiment of the present invention, FIG. 6 is a circuit diagram of a conventional example, FIG. 7 is a diagram showing the resonance characteristics of the same as above, and FIGS. 8 and 9 are operational waveform diagrams of the same as above. , FIG. 10 is a circuit diagram of another conventional example, and FIG. 11 is an operation waveform diagram of the same. (2) is a DC power supply, Q, , Q2 are switching elements, L is an inductor, C + and C2 are capacitors, RL is a discharge lamp, B is a lighting discriminator, and OSC1 to OSC3 are oscillators.

Claims (1)

[Claims] (1) A series circuit of first and second switching elements that do not block reverse current is connected in parallel to a DC power supply, and the first
and a drive circuit that alternately drives the second switching element on and off, and includes a resonant circuit that includes an inductor, a capacitor, and a discharge lamp and is excited by the voltage obtained at the connection point of the first and second switching elements. The discharge lamp lighting device includes a lighting discriminator that determines whether the discharge lamp is in the lighting state, and a switching element operating frequency when it is determined that the discharge lamp is not in the lighting state, and a resonance frequency of the resonant circuit. and a frequency control means that alternately switches between a first frequency higher than the resonant frequency and a second frequency higher than the resonant frequency and lower than the first frequency, the transition speed from the second frequency to the first frequency being controlled. is set to be sufficiently slower than the transition speed from the first frequency to the second frequency.