WO2002104081A1

WO2002104081A1 - Electronic ballast

Info

Publication number: WO2002104081A1
Application number: PCT/DK2002/000413
Authority: WO
Inventors: Remus Teodorescu
Original assignee: Aalborg Universitet
Priority date: 2001-06-19
Filing date: 2002-06-18
Publication date: 2002-12-27
Also published as: DE10129453A1

Abstract

The invention relates to an electronic ballast for a fluorescent tube (2), comprising a control device (9) and two step-up controllers (S1, S2) that share the same line branch (8). The aim of the invention is to be able to effect a pre-heating in a ballast of the aforementioned type. To this end, a switch (Q1) is placed in the line branch (8).

Description

Electronic ballast

The invention relates to an electronic ballast for a fluorescent tube, which has a control device and two step-up converters, which have a common line branch. Furthermore, the invention relates to a method for controlling an electronic ballast for a fluorescent tube, which has a control device and a switch bridge, the switch bridge including both an inverter and a step-up converter function.

Such a ballast is known from Tao, F. and Lee, F. C. : "An Interleaved Single-Stage Power-Factor-Correction Electronic Ballast" in IEEE 2000 as well as Tao, F. and Lee, F. C: "A Critical-Conduction-Mode Single-Stage Power-Factor-Correction Electronic Bal- last "in IEEE 2000, conference paper APEC 2000, February 6-10, 2000, New Orleans, USA.

In this known ballast, a special topology is described which assigns two functions to the switches in the half bridges of the inverters. One function is the normal inverter function. The other function is the power factor correction function via the step-up converter. Two different options are discussed: on the one hand the "transition rαode" (TM) and on the other hand the "discontinuous conduction mode" (DCM). In the first case, the current in the step-up converter rises again immediately after the zero crossing, while in the second case a longer pause is permitted in which the current remains at zero.

Such step-up converter topologies have several advantages. One of these is the reduction of the ripple current, which leads to smaller input filters at the input of the rectifier. On the other hand, the size of the step-up converter is reduced to a not inconsiderable degree.

However, such an electronic ballast has the disadvantage that it is difficult to preheat the fluorescent tube, as required by newer standards (IEC 929). The aim of such preheating is to extend the life of the fluorescent tube. The known ballast mainly works satisfactorily if the load is supplied with constant power. A reduction in the le Power supply to the fluorescent tube before the ignition is basically not possible.

The preheating of fluorescent tubes is known per se, see for example US Pat. No. 5,565,740 A. Here an electronic ballast for discharge lamps with hot cathodes is shown, in which the two functions (step-up converter and inverter) are designed in separate stages. You can therefore control each stage individually and thereby also preheat the fluorescent tube.

Another ballast is from "Improve ent on Component Stresses of Single-Stage Electronic Ballasts", Wu et al. , Industry Applications Conference

1999, Conference record of the 1999 IEEE, vol. 1, 1999, pages 285-292.

The object of the invention is to implement preheating with a ballast which has a step-up converter with a double function.

This object is achieved in an electronic ballast of the type mentioned in that a switch is arranged in the line branch.

With this switch, it is possible to interrupt the common line branch of the two step-up converters, which is provided in addition to the common output of the step-up converter. As soon as the common line branch has been interrupted, a current flow in this line branch is no longer possible. In this case, the two switches, which otherwise the two Implement a boost converter that no longer fulfills the boost converter function. Only the inverter function remains. The inverters then generate an alternating current from the direct current of the input rectifier, with which the cathodes of the fluorescent tube can be heated. The function of the step-up converter is given up. However, this is permissible because this functional interruption is only valid for a short period of time and in this period only the preheating of the cathodes of the fluorescent tube is actually necessary.

The switch is preferably designed as an optically controlled semiconductor switch. Such a switch can be designed, for example, as an optotriac, which is controlled by an LED (light-emitting diode). Such a combination of LED and optotriac is very reliable and inexpensive to implement.

The control device preferably adjusts a switching frequency of the switches of the step-up converters to a switching state of the switch. As soon as the switch has interrupted the common power branch of the step-up converters, the frequency for controlling the step-up converters, more precisely the switch of the step-up converters, can be set to a value which is favorable for the preheating of the cathodes. This is usually associated with an increase in the switching frequency compared to the operating frequency. This is possible because the switches only have to perform the inverter function and no longer the step-up function. In inverter and preheating mode, the resonant circuit, which is also called "resonant tank", has a first resonance frequency, while in operation a second resonance frequency results due to the heating. This second resonance frequency is used to ignite the fluorescent tube.

The control device preferably has a timer which determines the opening state of the switch. It can be used to specify a fixed time for preheating the fluorescent tube. Such a timer is very easy and inexpensive to implement. Basically, it only requires knowledge of the time it takes to preheat the fluorescent tube.

Alternatively, the control device can have a test device which actuates the switch depending on the state of the fluorescent tube. This embodiment has the advantage that it is much more precise and works with sufficient reliability even under operating conditions in which the fluorescent tube is briefly switched off and then on again, for example. In this case it can be assumed that the cathode of the fluorescent tube is still hot and requires a much shorter preheating period. If you use a fixed preheating period in this case, you can overheat the cathodes and thus shorten the life of the fluorescent tube.

It is particularly preferred here that the test device has a comparator which compares a first value which is influenced by the cathode current of the fluorescent tube with a second value which is influenced by the voltage at a cathode. The current through the cathode and the voltage across the cathode provide information about the current ohmic resistance of the cathode. The ohmic resistance of the cathode is also a measure of the temperature at the cathode. You can assign a temperature to each ohmic resistance in certain areas. If the temperature is "right" _/ is at the right value, you can stop the preheating and start the lighting of the fluorescent tube.

Preferably, a first value determination circuit is provided which uses a resonance inductor as part of a transformer. The resonance inductance is in any case present in the tank mentioned above, that is to say in the resonance circuit, which usually additionally has a capacitance parallel to the fluorescent tube. It is now relatively easy to arrange in parallel to this resonance type having a secondary winding to form a transformer .. _'De transformer detects the current through the cathode then, without galvanic coupling.

The transformer preferably has a secondary winding, to which a capacitance and an ohmic resistor are connected in parallel, a diode being arranged between the capacitance and the secondary winding. The diode has the effect that the current through the secondary winding can only flow in one direction, so that a DC voltage can be tapped off at the ohmic resistor and can be supplied to the comparator.

The transformation ratio n of the transformer is preferably determined according to the following relationship: n

where f _{ph is} the preheating frequency, L _{r is} the resonance inductance, R _{c is} the ohmic resistance of the cathode when cold and γ is an empirical constant. Usually γ is in the range from 4 to 4.25. In this case, a voltage signal U _{g is obtained} which is equal to γ • Rc • I _f , where I _{f is} the current through the cathode. In this way it is relatively easy to obtain a reliable assessment variable for preheating the cathode.

A determination circuit for the second value preferably has a capacitance parallel to the cathode, a diode being arranged between the cathode and the capacitance. With the capacitance, the voltage value can be easily determined via the cathode, the diode ensuring that the current can only flow into the capacitance in one direction. The diode therefore provides a kind of peak rectification. A DC voltage value is also obtained, which can be fed directly to the comparator.

The object is also achieved by the method, specifically by overriding the step-up converter function during a preheating phase.

As explained above, the resonance frequency can be set to a favorable value. The invention is described below on the basis of preferred exemplary embodiments in conjunction with the drawing. Show here:

1 shows a first embodiment of a ballast,

2 shows a second embodiment of a ballast and

3 shows a schematic illustration of a test device.

1 shows a ballast 1 for a fluorescent tube 2, which has a rectifier 3 and a switch bridge 4. The switch bridge 4 has two branches, each with a semiconductor switch S1, S2. A free-wheeling diode, which is known per se and is not shown in detail for reasons of clarity, is connected in parallel to both switches S1, S2.

Switch bridge 4 has two functions. It serves on the one hand as a step-up converter for the fluorescent tube 2, and on the other hand also as an inverter for the fluorescent tube 2.

The fluorescent tube 2 is connected to a common center tap 5 of the switch bridge 4. The other ends of the two branches of the switch bridge 4 and also the other ends of the fluorescent tube 2 are connected to ground 6, optionally with the interposition of a capacitance Cd. The fluorescent tube 2 is part of a resonance circuit or “tank”, which is formed by the series connection of an inductor Lr and a parallel connection of the fluorescent tube 2 and a capacitor Cr. If necessary, a further capacitance Cc can be arranged at the output of the switch bridge 4, ie in series with the center tap.

A smoothing capacitor Cf is connected in parallel with the rectifier. In parallel to this smoothing capacitor Cf is a capacitive voltage divider Cdl, Cd2, the center tap 7 of which is connected to the center tap 5 of the switch bridge 4 via a line branch 8. This line branch 8 is common to both switches S1, S2 of the switch bridge 4 and thus the two step-up converters.

A switch Q1, which can interrupt the line branch 8, is arranged in the line branch 8. The / switch Ql is designed as an optically controlled semiconductor switch, for example as an optotriac, which can be controlled by an LED (light-emitting diode) D3. The LED D3 is controlled by a control device 9, which also controls the ignition signals for the semiconductor switches S1, S2.

The rectifier is connected to the switch S1 via a diode Dl and an inductance Lbl. The switch S2 is again connected to the rectifier 3 via an inductor Lb2 and a diode D2.

When the fluorescent tube 2 is to be put into operation, a voltage Vin, for example the mains voltage, applied to the rectifier 3. The control device 9 then begins to preheat the fluorescent tube 2, more precisely the cathodes 11, 12 of the fluorescent tube 2. For this purpose, the switch Q1 is first opened by activating the light-emitting diode D3. If switch Q1 is opened, this means that the common return path of the two boost converter cells is interrupted, as a result of which the boost converter function is interrupted.

The next step is to apply a 74 kHz control frequency (preheating frequency f _ph ) to the control electrodes of switches S1 and S2. This frequency can be determined according to the following relationship:

I _{ph is} the preheating current specified by the IEC 929 standard, V _d is the DC voltage, R _{f is} the ohmic resistance of the cathodes 11, 12, Z ₀

(= V (Lr / Cr)) the characteristic impedance and ω ₀ the resonance frequency of the resonant circuit. The frequency ω can be isolated from this equation and the preheating frequency f _ph (ω = 2πf _ph ) can be determined therefrom. If each electrode, ie each cathode 11, 12, has an ohmic resistance R _f of approximately 10 Ω, the preheating current is approximately 500 A.

The switches Sl and S2 are opened and closed in push-pull mode, ie if Sl is conducting, S2 is open and versa. The control frequency and the push-pull when controlling the switches S1 and S2 cause an alternating current to flow. This alternating current flows in an instant from the positive rail of the inverter 3, ie from Cdl and Lbl, through S1, Cc, Lr, the first lamp electrode 11, Cr and the second lamp electrode 12 to the negative rail and in another instant when S2 is on , from the minus rail (from Cd2 and Lb2) through the second lamp electrode 12, Cr, the first lamp electrode 11, Lr and Cc.

Accordingly, during the preheating period, the half-bridge inverter is fed directly by the rectifier and not by the step-up converters.

After the preheating time has elapsed, the control device 9 closes the switch Q1 and thus puts the step-up converter back into operation. The control frequency is changed by the control unit 9 to the ignition frequency and the fluorescent tube 2 starts with a peak voltage of 600 V. The time from preheating to starting is approximately 1 microsecond. After the start-up sequence has ended, the control frequency is changed to approximately 48 kHz, which is the constant frequency during the normal operation of the fluorescent tube 2. This

Frequency is higher than the preheating frequency. It is tuned to the resonance circuit Lr, Cr.

The voltage across each tube electrode, ie the cathodes 11, 12, is determined by the control frequency which is applied to the control electrodes of the switches S1 and S2. In the illustrated embodiment, the ignition frequency is 45.8 kHz. So that you have a low When tension is reached across the filaments, one must choose a frequency that differs from the resonance frequency. As mentioned above, a frequency of 74 kHz is chosen. which leads to a voltage across the fila end with a size of 4 V. The rest of the DC voltage, approximately 325 V, is split between Cc, Lr and Cr.

The preheating period can be set as a fixed period. In many cases, however, it is advantageous to make the preheating dependent on a so-called "gamma ratio", that is to say the ratio Rh / Rc (the resistance of the warm cathode divided by the resistance of the cold cathode). In order to determine these values, a first determination circuit 13 and a second determination circuit 14 are provided, both of which are connected to the control device 9. The detailed structure of these two determination circuits 13, 14 is explained with reference to FIG. 3.

The first determination circuit 13 serves to determine a first value which is influenced by the cathode current and which is preferably proportional to the cathode current. The first determining device 13 has a transformer 15, the primary side of which is formed by the inductance Lr of the resonant circuit and the secondary side Lt2 of which is connected in parallel with a capacitor C1 and an ohmic resistor R1. A diode Dt is arranged between the capacitor C1 and the secondary side Lt2, which allows current to flow through the secondary side Lt2 only in one direction. A DC voltage U _g can therefore be taken off via the resistor R1. This DC voltage U _ς is a rator 16 supplied. The comparator 16 is part of a test device 17, which in turn forms part of the control device 9.

The second determination circuit 14 has a capacitor C2, which is connected in parallel with the second cathode 12 of the fluorescent tube 2. An ohmic resistor R2 is connected in parallel with the second capacitor C2. A diode Dt2 is arranged between the cathode 12 and the capacitor C2, so that here too a current flow in the loop, which contains the capacitor C2 and the cathode 12, is only possible in one direction. A voltage V can then be tapped via the resistor R2, which voltage is likewise fed to the comparator 16.

The comparator 16 generates a logic signal P which indicates whether the gamma value has reached its optimal value (typically in the range from 4 to 4.25). This is a sign that the cathodes 11, 12 of the fluorescent tube 2 have been preheated sufficiently.

If a secondary winding Lt2 with a small inductance is used, then a voltage is obtained which is proportional to the cathode current If, if the transformation ratio n is chosen according to the following relationship:

2π-fp ^ h - L. n = - r -R _e where f _{ph is} the preheating frequency, L _r the resonance inductance, R _c the ohmic resistance of the cold cathode and γ an empirical constant. In this case, a signal U _g = γ • R _c • I _{f is obtained} at the output of the first determination circuit 13. Since the second determination circuit 14 generates a signal V that is proportional to the cathode voltage at the cathode 12, the two signals in the analog comparator 16 are compared with each other. Initially, when cathode 12 is still cold, V is less than γ • R _c • I _f and signal P will remain low.

When the cathode resistance has reached its desired "hot" Rh value, the signal P goes high. It indicates that the ignition procedure can be started. The signal P then triggers an actuation of the light-emitting diode D3, which reverses the switch Q1, so that the line branch 8 becomes continuous.

If the control has been implemented digitally, the comparison function can of course also be controlled by software.

The exemplary embodiment in FIG. 1 uses the “discontinuous conductive mode” DCM mentioned at the beginning.

Instead, a "transition mode" (TM) can also be used, as shown in FIG. 2. The difference is that the inductors Lbl, Lb2 from the plus rail and the minus slide are combined in one inductor Lb, which is arranged in the common line branch 8. Otherwise, the circuit structure is unchanged. In this case, the preheating frequency is 74 kHz, the ignition frequency is 45.8 kHz and the operating frequency is 48 kHz.

In both exemplary embodiments, the ballast load is shown as a series resonance parallel load. Of course, other types of loads can also be used, such as parallel resonance parallel loads and the like. a.

Claims

claims

1. Electronic ballast for a fluorescent tube, which has a control device and two step-up converters, which have a common line branch, characterized in that a switch (Ql) is arranged in the line branch (8).

2. Ballast according to claim 1, characterized in that the switch (Ql) is designed as an optically controlled semiconductor switch.

3. Ballast according to claim 1 or 2, characterized in that the control device (9) tunes a switching frequency of the switches (Sl, S2) of the step-up converter to a switching state of the switch (Ql).

4. Ballast according to one of claims 1 to 3, characterized in that the control device (9) has a timer which determines the opening state of the switch (Ql).

5. Ballast according to one of claims 1 to 3, characterized in that the control device (9) has a test device (17) which actuates the switch (Ql) in dependence on the state of the fluorescent tube (2).

6. Ballast according to claim 5, characterized in that the test device (17) has a comparator (16) which compares a first value which is influenced by the cathode current of the fluorescent tube (2) with a second value which depends on the voltage a cathode (11, 12) is affected.

7. Ballast according to claim 6, characterized in that a determination circuit (13) is provided for the first value, which uses a resonance inductance (Lr) as part of a transformer (15).

8. Ballast according to claim 7, characterized in that the transformer (15) has a secondary winding (Lt2), a capacitance (Cl) and an ohmic resistor (Rl) are connected in parallel, a diode (Dt2) between the capacitance (Cl ) and the secondary winding (Lt2) is arranged.

Ballast according to claim 8, characterized in that the transformation ratio n of the transformer is determined according to the following relationship:

l * -f _PH - L _r n rR _c

where fp _{h is} the preheating frequency, Lr is the resonance inductance, Rc is the ohmic resistance of the cathode (11, 12) in the cold state and γ is an empirical constant.

10. Ballast according to one of claims 6 to 9, characterized in that a determination circuit (14) for the second value has a capacitance (C2) parallel to the cathode (12), a diode (between cathode and capacitance (C2)) Dt2) is arranged.

11. A method for controlling an electronic ballast for a fluorescent tube, which has a control device and a switch bridge, the switch bridge including both an inverter and a step-up converter function, characterized in that the step-up converter function during a preheating phase is overridden.