EP0936845B1 - Device and process for lighting and operating a fluorescent lamp - Google Patents

Description

The present invention relates to ignition circuits and fluorescent tube supply, and a method igniter and fluorescent tube supply.

Generally, a fluorescent tube must be powered at high frequency, for example at frequencies of the order from 10 to 100 kHz. In addition, it must receive alternating voltages. or in particularly intense pulses in the period initial to cause its priming. These impulses must reach voltages of the order of 1000 to 3000 volts. In a way general, to produce high voltages at high frequency, the fluorescent tube is associated with a resonant network made up of inductors and capacitors, this network being connected to a continuous or alternative power rectified through of switches controlled so as to periodically energize the resonant network.

Creation of a priming and supply circuit of a fluorescent tube poses problems for the realization of each of the elements of the system.

Regarding the resonant circuit, one of the constraints is that the cost of the elements is high, and in particular the cost of capacitors having to withstand very high voltages high and inductances caused to let a strong current pass all the more so as the value of these components is high.

Regarding the switching circuit, it must, for reasons of economy, understand as few as possible of switches and, preferably, all of these switches must be able to be produced on a monolithic silicon substrate. In practice, half-bridge systems are often used because they impose lower stress withstand stresses but they have the disadvantage of requiring at least two sets of monolithic switches.

Regarding the circuit control system switching, it should be as simple as possible and present a low consumption.

It is therefore clear that many compromises must be made. made to provide an optimal priming system and fluorescent tube supply, reducing the number of components and the cost of the system.

WO-A-9001248 and DE-A-4217822 describe examples of priming and feeding systems for fluorescent tubes.

It is an object of the present invention to provide a optimized fluorescent tube ignition and supply circuit.

To achieve this general object, the present invention provides a device for igniting and supplying a fluorescent tube, comprising a resonant system connected to the tube, this system having a first resonant frequency when the tube is primed and at least second and third resonant frequencies when the tube is not initiated, the third resonant frequency being higher than the first and second resonant frequencies; a power circuit rectified connected to the resonant system; a switch in series between the power supply and the resonant circuit; a first detector for control the switch on opening when the current supplied by the supply exceeds a determined threshold; and a second detector for command the switch to close at each zero crossing of the voltage on a node of the resonant system and each time a minimum of this voltage.

According to an embodiment of the present invention, the resonant system comprises a first capacitor and a first inductor connected in series across the tube, and a second capacitor and a second inductor connected in parallel to the terminals of the tube, the second capacitor having a lower capacity than the first capacitor.

According to an embodiment of the present invention, the second detector includes a branch circuit whose output is connected to a zero crossing detector indicating zero crossings in a specific direction.

According to an embodiment of the present invention, the second detector includes a transistor whose emitter is connected to the node of the resonant system via a capacitor and whose transmitter is connected to the base via resistance, the base being connected to ground by through a clean diode to allow a current of mass control to the node through the resistance to bias the transistor to the conduction, and the time constant is much less than the signal period of highest frequency resonance that one wishes to detect.

According to an embodiment of the present invention, the switch includes a power MOS switch whose grid is controlled at opening and closing, in series with a bipolar transistor whose base is permanently polarized.

According to an embodiment of the present invention, the circuit includes a supply node connected to ground via a storage capacitor, this node on the one hand connected to the high power through a high value resistor, else leaves at the base of said bipolar transistor to receive one load destocking current at each opening of this transistor, and to the capacitor of the second detector to receive the excess charge.

The present invention also provides a method of initiation and supplying a fluorescent tube comprising the steps of providing a resonant system connected to tube terminals, this system having a first resonant frequency when the tube is primed and at least second and third resonance frequencies when the tube is not primed, the third resonant frequency being higher than the first and second resonant frequencies; connect this resonant system to a supply circuit rectified by through a controlled switch; detect current into the switch and open the switch whenever this current exceeds a determined threshold; and detect the voltage on a node of the resonant system and automatically adapt the closure from the switch to the highest resonant frequencies of the resonant circuit.

According to an embodiment of the present invention, the highest frequency detection step of the circuit resonant consists in detecting the minima of the voltage present on a node of the resonant circuit and the zero crossings of this voltage.

la figure 1 est un schéma sous forme de blocs d'un circuit de démarrage et d'alimentation de tube fluorescent selon la présente invention ;

la figure 2 représente l'allure de signaux apparaissant dans un circuit résonant ;

la figure 3 représente un mode de réalisation plus détaillé du circuit de la figure 1 ;

la figure 4 représente un exemple détaillé de réalisation du circuit de la figure 3 ; et

les figures 5A à 5C représentent des variantes de réalisation du circuit résonant.

These objects, characteristics and advantages, as well as others of the present invention will be explained in detail in the following description of particular embodiments given without limitation in relation to the attached figures among which:

Figure 1 is a block diagram of a fluorescent tube start-up and supply circuit according to the present invention;

FIG. 2 represents the appearance of signals appearing in a resonant circuit;

Figure 3 shows a more detailed embodiment of the circuit of Figure 1;

FIG. 4 represents a detailed example of an embodiment of the circuit of FIG. 3; and

Figures 5A to 5C show alternative embodiments of the resonant circuit.

According to a characteristic of the invention, the network resonant associated with the fluorescent tube according to the present invention, has a first resonant frequency when the tube is passing, and has several resonant frequencies including one at least is higher than the first resonant frequency when the tube is not yet primed (and it is substantially equivalent to an open circuit). We will note in the particular example below and it will generally be noted that the fact of operating at a higher frequency for high voltages data allows the capacitors to support the high voltages may have lower values and also causes the currents in the inductors of the network will be weaker. This therefore allows the use of lower cost capacitors and inductors.

More particularly, Figure 1 shows a tube fluorescent 1 in which it was assumed that there was no preheating of electrodes. This fluorescent tube is associated with a resonant network comprising capacitors C1 and C2 and inductors L1 and L2. The inductance L1 and the capacitor C1 are connected in series to the tube terminals. Inductance L2 and the capacitor C2 are connected in parallel to the terminals of the tube. A terminal of a continuous power source, for example a rectified alternating supply Vdd, is connected to the terminal of the tube connected to a terminal of capacitors C1 and C2 and of inductance L2. The connection point of capacitor C1 and inductance L1 constitutes a node N1 of the circuit.

C1 = 1 nF,

L1 = 6,4 mH,

L2 = 25 mH, et

C2 = 300 pF.

We will consider below a particular example, indicated only by way of example, in which the applied voltage is the rectified mains voltage (220 V) and where the elements of the resonant network have the following values:

C1 = 1 nF,

L1 = 6.4 mH,

L2 = 25 mH, and

C2 = 300 pF.

Those skilled in the art will appreciate that, once a tube is primed, it has a low impedance, for example a resistance of the order of 500 Ω. Since the tube of figure 1 is arranged in parallel with the capacitor C2 and the impedance L2, these latter elements are amortized and no longer have any influence on the resonant system once the tube is primed. The network resonant then reduces substantially to capacitor C1 and to inductance L1 which then define the oscillation frequency (of the order of 90 kHz in the case of the particular example above).

When the tube is not primed, we can consider that the network has two main resonant circuits. A first resonant circuit consists of inductors L1 and L2 in series with capacitor C1. This first resonant circuit will have a resonance frequency of the order of 28 kHz in the case of the particular example above. A second resonant circuit includes inductance L1 in series with capacitors C1 and C2. The resonant frequency of this second resonant circuit will be of the order of 126 kHz in the case of the particular example above. This shows that the network will have at least two frequencies resonance when the tube is not primed and gives approximate orders of magnitude of the resonant frequencies for indicate that there will be a clearly high resonant frequency higher than the resonant frequency in the primed state and a low resonant frequency. So we get, when the circuit oscillates, a wave of complex shape comprising at least the superimposition of a high frequency signal and a frequency signal low.

Node N1 is connected to the second power supply terminal GND (commonly ground) via a switch SW and is connected directly to the GND terminal by a diode in reverse D1.

The SW switch is controlled by the Q output of a flip-flop 10 set to 1 by a starting circuit 11.

Flip 10 reset input is connected to a current detection circuit 12 in the switch SW, this detection circuit providing an output signal when the current exceeds a determined threshold, for example a value of 200 milliamps.

The clock input of flip-flop 10 is controlled by a detector circuit 14 which provides an active signal on the input CLOCK, i.e. a signal passing from a low state to a state high when the voltage on node N1 remains at zero after having been positive or when this tension goes through a minimum. This allows, as will be seen below, to control the switch on the highest frequency among the resonant frequencies above.

Figure 2 shows by way of example the voltage on the node N1. We assume that, at time t1, the switch SW is closed. It opens as soon as the current flowing through it exceeds a threshold and the voltage at node N1 increases and has a waveform relatively complex illustrated between instants t1 and t2, consisting in particular of the superposition of resonance frequencies above and below. At time t2, this tension goes through zero and detector 14 provides a signal on CLK input of flip-flop 10 to close the SW switch. AT time t3 when detector 12 has detected a higher current at 200 milliamps, the switch opens again. We find then a complex waveform and there will necessarily come a moment (during this period, the previous tl-t2 period, or a later period) where the superimposition of high frequencies and bass will cause, at time t4, that this waveform goes through a minimum. This minimum corresponds to a low value of the high frequency component. At this time, the detector 14 provides a rising edge on the CLOCK input of flip-flop 10. The output Q of flip-flop 10 then applies a closing signal on the SW switch control terminal. Starting from moment, there is hooking on the high frequency. And the switch opens and closes substantially at this frequency, the opening occurring whenever the current in the switch exceeds a value of 200 milliamps and the closure occurs producing every time we go through a minimum or a zero voltage crossing at high frequency.

Figure 1 also shows an embodiment simplified detector 14. This detector comprises, between the node N1 and ground (GND), a capacitor C3 and a resistor R3 whose connection point N2 is connected to an input of a comparator 16. The other input of the comparator is connected to a negative reference voltage. This negative reference allows cause a positive edge on the CLK input of flip-flop 10 when the voltage on node N1 remains at 0 (or at -0.6 volts at due to the presence of diode D1) after being positive. The time constant R3C3 is chosen much less than the period of the signal corresponding to the most resonant frequency high. The fixture works like a diverter and the voltage at node N2 goes through zero with each change in slope of the voltage on node N1. Comparator 16 provides a transition from a high state to a low state when the voltage on node N1 goes through a maximum and from a low state to a high state when it goes through a minimum. Toggle 10 only provides a signal on its Q output only during transitions from a low state to a high state on its CLOCK entry. We therefore obtain the control signal for sought switching that automatically hooks onto the signal at the highest frequency among the components of resonant circuit signal.

In addition, examples have been indicated previously. numerical values of capacitors C1 and C2. We will remember that the capacitor C2 has a much lower capacity than the capacitor C1. If its capacity is, for example, three times lower, the voltage across it will be about three times as much strong, i.e. if the voltage across the capacitor C1 is of the order of 300 volts, the terminals of the capacitor C2 of peak-to-peak voltages of the order of a thousand volts, sufficient to trigger the fluorescent tube.

After a certain number of switch operations SW at the high frequency, the fluorescent tube will strike and, as as indicated above, only the capacitor C1 and the inductance L1 will then be active in the resonant circuit. Therefore the detector 14 will automatically adjust to the new frequency and will provide switching pulses from the SW switch to each zero crossing of the alternating voltage corresponding to the resonant frequency of the L1-C1 network.

FIG. 3 represents an exemplary embodiment more detailed view of the circuit in figure 1. In this figure, the same elements as those in Figure 1 are designated by the same digital references.

The resonant system associated with tube 1 is identical to that of figure 1.

The SW switch is made by cascode mounting a bipolar transistor 20 and a MOS transistor 21. Such components can be made in monolithic form in a single chip, for example in integration technologies bipolar-MOS developed by the company SGS-THOMSON. The collector of transistor 20 is connected to node N1, its emitter to drain of transistor 21, and its base at a node N3 on which is available a low supply voltage (+ Vcc). The drain of transistor 21 is connected to ground via a measuring resistance R4. The gate of transistor 21 is connected to the output Q of the flip-flop 10. The transistor 20 is permanently polarized in the passing state and a current does not cross it effectively only when the MOS transistor 21 turns on. The role essential of bipolar transistor 20 is to limit the voltage across the MOS transistor 21 which sees only the voltage emitter of this transistor 20 (substantially equal to the voltage VDC). Indeed, it is technologically easier to achieve a bipolar transistor supporting a high voltage than a transistor MOS supporting high voltage.

The current detector 12 includes a resistor R4 whose voltage (node N4) is applied to the base of a transistor NPN 23 whose transmitter is connected to ground and the collector to the supply node N3 via a resistor R5. The collector voltage of transistor 23 is applied to the reset input R of flip-flop 10. Thus, as soon as the voltage across resistor R4 exceeds base-emitter voltage of transistor 23 (approximately 0.6 volts), this transistor becomes passing and a low level appears on its collector. The low level is applied via an inverter (a first entry of a NAND gate 25) at entry R. If you want that the MOS transistor 21 opens as soon as a current of the order of 200 milliamps crosses it, we will choose for resistance R4 a value of 3 Ω.

The circuit 14 for detecting passage through a minimum or by zero of the voltage on node N1 includes the capacitor C3, a first terminal of which is connected to this node N1 and the second terminal is connected to ground via a capacitor C4. We denote by the reference N5 the point of connection of capacitors C3 and C4. N5 node is connected to node N3 via a diode D2. In addition, the circuit 14 includes a resistor R3 connected between base and emitter a transistor 27 whose emitter is connected to node N5 and whose the collector is connected to node N3 via a resistance R6. Earth is connected to the base of transistor 27 by via a diode D3 and to the collector of this transistor via a diode D4. If node N5 is more positive as -1.2 V, transistor 27 is blocked. If node N5 becomes more negative than -1.2 V, i.e. a current flows through capacitor C3 from node N5 to node N1, this current flows from ground through diode D3 and the resistance R3 towards node N5 and the voltage which develops at resistance R3 causes the conduction of the transistor 27. Its collector then passes from the voltage level of the node N3 (high level) at the voltage level of node N5 (level low). This transition causes the appearance of a signal on the CLK entry. The same phenomenon occurs when the voltage of the node N1 remains at zero after being positive. In this case resistor R3 blocks transistor 27 after current cancellation in capacitor C3.

The starting circuit 11 firstly comprises a resistor R7 and a capacitor C7. Resistor R7, connected between the voltage Vdd and the node N3, charges the capacitor C7, connected between node N3 and ground, as soon as a voltage is applied to the terminal Vdd and positively polarizes the node N3. A Zener Z diode sets the maximum voltage level. As soon as the capacitor C7 is sufficiently charged, a circuit comprising resistors R8, R9, R10, R11, R12, R13, NPN transistors 29 and PNP 30, and a capacitor C8 connected in the illustrated manner, provides a signal on the setting input to 1, S, of the flip and on the input R of it via the door 25 above. The voltage on node N3 is applied at entry D of the scale. As long as the voltage on node N3 is too weak, transistors 29 and 30 are blocked and the flip-flop 10 is kept blocked by the signal applied to the carries 25. When that the tension on the node N3 crosses the trigger threshold of transistors 29 and 30, the capacitor C8 applies a pulse to the S input of the flip-flop.

In addition, the signal on the output Q flip-flop 10 is applied via a capacitor C9 and a resistor R14 at the base of transistor 23 to reset it to zero with a certain delay. The exit Q is used to inhibit the operation of transistor 23 each time the switch SW is turned on. Indeed, the switch SW can be turned on while there is a high voltage across its terminals, which induces a lot of current in the resistor R4. The capacitor C9 makes it possible to apply a negative pulse to the base of the transistor 23, which avoids re-blocking the flip-flop 10 just after setting it to 1.

One aspect of the present invention also resides in the low supply voltage generation mode on the node N3. An initial charging step has been indicated via of resistance R7. The present invention provides two other means of supplying this DC voltage. The first is that whenever the transistor 20 opens as a result of blocking of the MOS transistor 21, the charges stored in this transistor will be eliminated towards the node N3 by through a resistor R15. The second uses any excessive energy on the capacitor C3 which is discharged by through the diode D2 in this node N3. So we use for this load basically voltages and loads which otherwise would be lost. This helps maintain sufficient tension on node N3 during all operating phases retaining a resistance R7 of very high value (by example 1 MΩ) to limit unnecessary consumption of the circuit.

Figure 4 shows a detailed embodiment of the present invention. In this figure, we have represented some additional components compared to those of the Figure 3 intended to ensure proper operation of the circuit. In particular, the Q output of flip-flop 10 is applied to the grid of the switching MOS transistor 21 via a amplifier circuit and the output voltage of the supply circuit is applied via two inverters. The usefulness of the other elements added will become clear to one skilled in the art. In addition, this figure has indicated the value and / or type of each component used in a mode of particular achievement. These values, given by way of example, will be considered part of this description.

Thus, the present invention provides a simple system switch control for automatic adjustment on the highest frequency of a likely resonant system to oscillate at several frequencies.

The present invention is susceptible of various variants and modifications which will appear to those skilled in the art. In particular, it should be noted that the numerical values indicated were used only as an example. In addition, a particular type of resonant circuit. Various other structures resonant circuit may be used, the important being that this circuit has a frequency of high resonance which is automatically inhibited once that the tube is primed. In addition, a electrode heating system, and possibly modify the resonant circuit accordingly.

Examples of variants of the resonant circuit are illustrated in FIGS. 5A, 5B and 5C, the variant of FIG. 5C providing for electrode heating.

Claims (8)

1. A device for starting and supplying a fluorescent tube, including:

   a resonant system (C1, C2, L1, L2) connected to the tube, this system having a first resonance frequency when the tube is started and at least second and third resonance frequencies when the tube is not started, the third resonance frequency being higher than the first and second resonance frequencies;

   a rectified supply (Vdd, GND) circuit connected to the resonant system;

   a switch (SW) in series between the supply and the resonant system;

   a first detector (12) for controlling the switch (SW) to turn off when the current provided by the supply exceeds a determined threshold; and

   a second detector (14) for controlling the switch (SW) to turn on for each transition through zero of the voltage on a node (N1) of the resonant system and for each transition through a minimum of this voltage.
2. A starting device according to claim 1, wherein the resonant system includes a first capacitor (C1) and a first inductance (L1) connected in series across the tube, and a second capacitor (C2) and a second inductance (L2) connected in parallel across the tube, the second capacitor (C2) having a lower capacity than that of the first capacitor (C1).
3. A device according to claim 1, wherein the second detector (14) includes a shunting circuit (C3, R3), the output of which is connected to a zero detector (16) indicating transitions through zero in a determined direction.
4. A device according to claim 3, wherein the second detector (14) includes a transistor (27), the collector of which is connected to a node (N1) of the resonant system via a capacitor (C3) and the emitter of which is connected to the base via a resistor (R3), the base being connected to the ground via a diode (D3) for letting a control current run through from the ground to the node (N5) via the resistor (R3) to bias the transistor upon conduction, and wherein the time constant (R3, C3) is much lower than the period of the resonance signal having the highest frequency which is desired to be detected.
5. A device according to claim 1, wherein the switch (SW) includes a MOS power transistor (21), the gate of which is controlled to open and close, in series with a bipolar transistor (20), the base of which is constantly biased.
6. A device according to claims 1, 4, and 5, wherein the circuit includes a supply node (N3) connected to the ground via a storage capacitor (C7), this supply node being connected on the one hand to the high supply via a high value resistor (R7), on the other hand to the base of the bipolar transistor to receive therefrom a discharge current upon each opening of this transistor, and to the capacitor (C3) of the second detector to receive the excess charge therefrom.
7. A method for starting and supplying a fluorescent tube, including the following steps:

   providing a resonant system (C1, C2, L1, L2) connected across the tube, this system having a first resonance frequency when the tube is started and at least second and third resonance frequencies when the tube is not started, the third resonance frequency being higher than the first and second resonance frequencies;

   connecting this resonant system to a rectified supply circuit via a controlled switch (SW);

   detecting the current in the switch and opening the switch each time this current exceeds a determined threshold; and

   detecting the voltage on a node of the resonant system and automatically adapting the closing of the switch to the highest of the resonance frequencies of the resonant circuit.
8. A method according to claim 7, wherein the step of detecting the highest frequency of the resonant circuit consists of detecting the minima of the voltage present on a node of the resonant circuit and the transitions through zero of this voltage.

Images