Electronic ballast system having emergency lighting provisions and electronic chip

[0001] This application is a continuation-in-part of pending application Ser. No. 10/025,318 which is now U.S. Pat. No. 6,784,624 filed Dec. 19, 2001.

[0002] This invention relates to electronic ballast systems, and, more particularly, to an electronic ballast system for operating fluorescent lamps with universal input (from 108V to 305V) and having a subsystem with a battery, along with a battery charger, and associated logic that provides for emergency lighting.

[0003] Electronic ballast systems for operating fluorescent lamps are well known and some are disclosed in U.S. Pat. Nos. 5,808,421, 6,031,342 as well as other patents. Electronic ballast systems typically convert a low frequency alternating current source having a relatively low frequency in the range from 50 to 60 Hz to a higher frequency typically in the range of 30–40 kHz. The conversion commonly involves a two stage process, wherein the ac oscillation having a frequency of 50 to 60 Hz is first rectified to a dc voltage and then this dc voltage is chopped at a higher frequency to produce alternating current in the frequency range of 30–40 kHz, which is used to excite the fluorescent lamp. The electronic ballast circuits advantageously perform the desired function for operating fluorescent lamps and reduce the energy consumption, compared to non-electronic ballast circuits, and especially compared to incandescent lamps. However, the conventional electronic ballast circuits typically employ a forced ignition for rapid start of the fluorescent lamps, but with this technique the voltage reaches as high as 2.000 volts. This unfortunately consumes and quickly destroys the cathodes of the fluorescent lamps. It is desired, however, that an electronic ballast circuit be provided that through a soft starting mode, energizing the lamp slowly, following a starting sequence controlled by an electronic chip, the voltage of the cathodes of the lamps only raises to 1.000 volts, and therefore it lengthens the operating life of the fluorescent lamps.

[0004] Electronic ballast circuits are typically selected to have parameters that operate for particular input frequency oscillation. For example, the electronic ballast circuit may have parameters selected so as to operate with the 110 volt, 60 Hz typically found in the United States, whereas other ballast circuits may have parameters selected to operate with 220 volts, 50 Hz typically found in European countries. It is desired to provide a ballast circuit that operates with the universal input covering the range from between 108 to 305 volts at a frequency range between 50–60 Hz.

[0005] Further, it is desired to provide an electronic ballast circuit that handles various types of fluorescent lamps such as, T2, T5, T8, T10, T12, 20W, 32W, 40W, 56W, 58W, 70W, linear, circular, 2D, twin tube, compact or U-shaped type fluorescent lamps.

[0006] Because fluorescent lamps have a lower energy consumption compared to incandescent lamps, they are extensively used in industrial and commercial environments, which commonly require emergency lighting. It is desired to provide for electronic ballast system for operating one or more fluorescent lamps and having a subsystem with a battery, along with a battery charger, and associated logic to provide for emergency lighting.

[0007] Electronic ballast systems that operate fluorescent lamps are commonly plagued by the disadvantage that they produce electromagnetic interference (EMI) and radio frequency interference (RFI). It is desired to provide for an electronic ballast system that reduces or even eliminates the EMI/RFI noise commonly produced by the electronic ballast systems.

[0008] Electronic ballast circuits commonly employ inductive loads, which act to lower the power factor, which, in turn, increases the consumption of current and, thereby, reduces the efficiency related to fluorescent lamps. It is desirable that an electronic ballast circuit be provided with a power factor correction circuit that allows for the creation of a power factor that approaches unity, thereby furthering the efficiency of the electronic ballast system.

[0009] The invention is directed to an electronic ballast system for operating one or more fluorescent lamps with universal input (from 108V to 305V) and having a subsystem with a battery, along with a battery charger and associated logic, that provides for emergency lighting. The electronic ballast system allows for the operation of fluorescent lamps with preheating their cathodes, as well as a circuit for power factor correction that allows electronic circuit to have a power factor and ballast factor that approaches unity.

[0010] The electronic ballast system comprises: (a) an EMI filter having an input connected to an electric surge and providing a filtered output; (b) a full-wave rectifier having an input connected to the output of said EMI filter and providing a first d.c. voltage; (c) a power factor correction circuit having an input connected to the output of the full-wave rectifier and providing a power factor regulated output; (d) a first inverter ballast circuit having an input connected to the output of the power factor converter; and (e) an electronic chip which limits the voltage increase in the cathode of the lamps, and serves to control the circuit and contribute to its safety. The inverter ballast circuit has a sweep frequency circuit for supplying an oscillating current to power a first fluorescent lamp without the need of preheating its cathode.

[0017] With reference to the drawings, wherein the same reference number indicates the same element throughout, there is shown in FIG. 1 an electronic ballast system 10 for operating one or more fluorescent lamps and having the capability of providing for emergency lighting.

[0018] The electronic ballast system 10 comprises an electromagnetic interference (EMI) filter 12 having input terminals L1 and N, connected to a source of current, and providing a filtered output on signal paths 14 and 16. The electronic ballast system 10 further comprises a full-wave rectifier VI having its inputs connected to signal paths 14 and 16 of the EMI filter 12 and providing a first rectified d.c. voltage on signal paths 18 and 20. A power factor correction circuit 22 has its inputs connected to signal paths 18 and 20 of the full-wave rectifier V1 and provides a power factor regulated output on signal paths 24 and 26.

[0019] The electronic power system further comprises a first inverter ballast 28A and preferably further comprises three additional inverter ballast circuits 28B, 28C, and 28D. Each of the inverter ballast circuits 28B, 28C, and 28D has their inputs connected to the signal paths 24 and 26 of the power factor correction circuit 22. Further, each of the ballast circuits 28A, 28B, 28C, and 28D has a sweep frequency circuit supplying an oscillating current to provide power to a fluorescent lamp into its operating or running mode with the rapid preheating the cathode of the fluorescent lamp.

[0020] The electronic ballast system 10 comprises a static switch connected to the inputs signal path 24, of the power factor correction circuit 22, and path 38 of the switching power supply circuit shown in FIG. 6, and connected to the output signal path 42 of the inverter ballast circuits 28A, 28B, 28C, 28D. The static switch serves to send the signal of +14V, coming from the power supply path 38, to the ballast circuits 28A, 28B, 28C, 28D and path 42 only when it is present and the supply voltage or the wall switch is turned on.

[0021] The electronic ballast system 10 has a capability of providing for emergency lighting from a subsystem comprised of a switching power and battery charger 30 and an emergency section 32.

[0022] The switching and battery charger 30 has a power supply having its inputs connected across hot and ground terminals of the EMI filter 12 and provides a second d.c. voltage having positive and negative potentials applied to signal paths 34 and 36, respectively.

[0023] The emergency section 32 has switching means K1 allowing connection to the signal paths 34 and 36 and has a plurality of contacts SC1 and SC2. The emergency section 30 further comprises a diode D16 having an anode and cathode with the anode connected to the positive potential on signal path 34. The emergency section 30 still further comprises a battery having positive and negative terminals with a negative terminal connected, by way of signal path 38, to the negative potential on signal path 36 and the positive terminal thereof connected to the cathode of the diode D16.

[0024] The emergency section further includes a second ballast inverter comprised of a half-bridge driver U4A, a half-bridge arrangement 40, and a high voltage transformer T2. The second inverter circuit has an input and output arranged by means of the plurality of the switch contacts SC1 and SC2 so as to interconnect the positive and negative terminals of the battery when the second d.c. voltage normally applied across terminal 34 and 36 is absent. The second inverter circuit has a sweep frequency circuit for supplying an oscillating current to a first fluorescent lamp with the rapid preheating the first fluorescent lamp in a manner similar to that of inverter ballast circuits 28A, 28B, 28C, and 28D. The second inverter ballast circuit further comprises means, to be discussed with reference to FIG. 5, for energizing signal paths 42 and 44 so as to excite the fluorescent lamp of the inverter ballast circuit 28A.

[0025] The electronic ballast system 10, more particularly, the EMI filter 12 is comprised of a plurality of elements as listed in Table 1, and may be further described with reference to FIG. 2.

[0026]

	TABLE 1
	Element	Typical Value/Type
	S11	Conventional fuse
	R1	Varistor
	C1	Film Capacitor (Interference Suppressor)
	C2	Film Capacitor (Interference Suppressor)
	C3	Film Capacitor (Interference Suppressor)
	L1 and L2	Common mode inductor

[0027] The purpose of the EMI filter 12 is to reduce or even eliminate the electro-magnetic interference (EMI) and radio interference (RFI) serving as sources of noise that would otherwise most likely be produced by the electronic ballast system 10. This electrical noise can interfere with the operation of televisions, radios, telephones, and similar equipment. Further, this electrical noise can be conducted through power lines which radiate and create disturbances in external equipment.

[0028] The EMI filter 12 is connected across the source of voltage which, because of benefits of the present invention, may vary from about 100 to 305 volts a.c. with a frequency variation between 50–60 Hz. The EMI filter 12 is comprised of the common mode inductors L1 and L2, wound around the same core and operatively connected to capacitors C1, C2, and C3 arranged as shown in FIG. 1, and provides RFI/EMI filtering of common mode noise. Additionally, the EMI filter 12 includes a fuse SI1 that provides for over current protection and a varistor R1 that provides protection against high voltage spikes. The EMI filter 12 provides a filtered output on signal paths 14 and 16, which is applied across a full-wave bridge rectifier V1 comprised of conventional diodes arranged as shown in FIG. 2.

[0029] The full-wave bridge V1 operates in a usual way to convert the filtered output of the EMI filter 12 into a d.c. voltage which is applied, via signal paths 18 and 20, across a leveling capacitor C4, which is part of the power factor correction (PFC) circuit 22 comprised of the plurality of elements given in Table 2 and arranged as shown in FIG. 2.

[0030]

	TABLE 2
	Element	Typical Value/Type
	C4	Electrolytic Capacitor
		(Smoothing Capacitor)
	C5	Film Capacitor
	C6	Film Capacitor
	C7	Electrolytic Capacitor
	C8	Electrolytic Capacitor
	C9	Electrolytic Capacitor
	R2	Resistor
	R3	Resistor
	R4	Resistor
	R5	Resistor
	R6	Resistor
	R7	Resistor
	R8	Resistor
	R9	Resistor
	R10	Resistor
	D1	Diode Fast Recovery
	D2	Diode Fast Recovery
	D3	Boost diode
	Q1	Controlled power switch
	T1 having winding	High Frequency Transformer
	T1A and T2A
	U1A	Power Factor Control

[0031] The purpose of the power factor correction circuit 22 is to derive a power factor for the electronic power system 10 that approaches unity. The power factor correction circuit 22 is preferred to be implemented into the electronic ballast system 10 because the electronic ballast system 10, as well as all known ballast circuits, uses inductors which tend to reduce the power factor as seen by the source of a.c. excitation supplying the electronic ballast system 10. This lowering of the power factor disadvantageously increases the consumption of power related to the fluorescent lamps and the magnetic ballast components. Typically this reduction in the power factor and related inductive disturbances creates a 40% increase in power consumption. The power factor correction circuit 22 of the present invention performs an active correction of the power factor typically bringing it almost to unity (0.98%) and accomplishes this correction by forcing its output present signal paths 24 and 26 to follow the average primary current of the a.c. supplying the electronic ballast system 10. Further, the power factor correction circuit 22 maintains a dc voltage of approximately 450 volts that is stabilized regardless of the fluctuation of the a.c. that may vary from 108 to 305 volts.

[0032] In general, and with simultaneous reference to FIGS. 1 and 2, the power factor correction circuit 22 comprises an inductor T1 having a first winding T1A having input and output terminals with the input terminal connected to the positive terminal 18 of the full-wave rectifier V1. The power factor correction circuit 22 further comprises a power switch Q1 having first (1) second (2) and third (3) electrodes with the first electrode (1) connected to the output terminal of the winding T1A. The power factor correction circuit 22 further comprises a diode D3 having an anode and a cathode with the anode connected to the output of the first winding T1A and a cathode connected to the positive terminal 24 of the second dc output generated by the power factor correction circuit 22.

[0033] Capacitive means comprised of capacitor C8 and C9, is arranged across the positive and negative terminals connected to signal paths 24 and 26. A P.F.C. controller U1A having an input and an output with the input (pin 3) connected across the positive and negative terminals present on signal paths 18 and 20 and is connected thereto by means of a network comprised of capacitors C4 and C5 and resistors R2 and R3 arranged as shown in FIG. 2. The P.F.C. controller U1A has means, including a pulse width modulation control, so that the controller U1A provides an output that varies in accordance with the primary current created by the full-wave rectifier V1. The controller U1A is connected to the second (2) electrode of the power switch Q1 and the P.F.C. controller U1A is also connected to the third electrode (3) of the power MOSFET switch Q1.

[0034] The P.F.C. controller U1A contains a wideband voltage amplifier used in an internal feedback loop, an overvoltage regulator, a quadrant multiplier having a wide linear operating range, a current sense capacitor, a zero current detector, a pulse width modulator (PWM) having associated logic circuitry, a totem-pole arranged MOSFET driver, an internal voltage reference, a restart timer, and an under voltage lockout circuit. The controller U1A is interconnected into the circuitry shown in FIG. 2 by means of its pins 1–8.

[0035] Pin 1 (IN) of the P.F.C. controller U1A serves as a voltage amplifier inverting input. This pin 1 is connected, via a resistive divider R9, R10, and R8, to the signal path 24. Pin 2 (COMP) of P.F.C. controller U1A serves as a voltage amplifier output and is the output of the error amplifier (and one of the two inputs to the internal quadrant multiplier). A feedback compensation network, contained in the P.F.C. controller U1A, reduces the frequency block gain to advantageously avoid an attempt of the P.F.C. controller U1A to control the output voltage ripple (120 Hz) carried on signal paths 24 and 26. This pin 2 is connected, via capacitor C6, to the ground connection.

[0036] Pin 3 (MULT) of the P.F.C. controller U1A serves as the second input to quadrant multiplier. Pin 3 is connected, through a resistive divider R2, R3, to signal paths 18 and 20. Pin 4 (CS) of the P.F.C. controller U1A serves as an input to the current sense comparator. This input (Pin 4) provides the instantaneous MOSFET current derived from power switch Q1 and which current is represented by a proportional voltage signal created across an external sense resistor R7. This proportional voltage signal is compared with the threshold set by the quadrant multiplier output, and when the resulting current therefrom exceeds the set value, the power MOSFET Q1 will be turned off by a reset signal provided by the quadrant multiplier and remains off until the next set pulse generated by the PWM latch of the quadrant multiplier.

[0037] Pin 5(TM) of the P.F.C. controller U1A serves as the input zero current detector. Pin 5 is connected, through a limiting resistor R5, to the auxiliary winding T1B of the inductor T having the primary winding T1A. Pin 5 provides a zero coil current and voltage function for the P.F.C. controller U1A which processes the inductor signal derived from auxiliary winding T1B and turns on the external MOSFET Q1 as the voltage at pin 5 crosses the threshold level set by the quadrant multiplier. Pin 6 (GND) of the P.F.C. controller U1A is the common reference of the circuitry of FIG. 2. Pin 7 (OUT) of the P.F.C. controller U1A is output of the totem-pole arranged MOSFET driver. This pin 7 serves to drive the external MOSFET Q1. Pin 8 (Vcc) of the controller U1A carries the supply voltage. This pin 8 is externally connected to a rectified diode D1 and a filter capacitor C7 as shown in FIG. 2.

[0038] The P.F.C. controller U1A operates within a voltage range of 108V to 305V and uses average current mode PWM control to provide line and load compensations. The P.F.C. controller U1A uses an optimum current control method. More particularly, the controller U1A uses an average current control implemented by using feed forward line regulation and a variable switching frequency. An oscillator, within the P.F.C. controller U1A, simultaneously turns on the MOSFET power switch Q1 and starts the ramp of the PWM current control to regulate the output present on signal paths 24 and 26.

[0039] The average inductor current present on pin 4 of the P.F.C. controller U1, is compared with a current reference generated by means of a current error amplifier of the quadrant multiplier. The P.F.C. controller U1A operates as an integrator, allowing the P.F.C. controller U1A to accurately provide an output that follows the current reference generated by the quadrant multiplier.

[0040] A so called “feed forward compensation” of the voltage present on signal paths 18 and 20 has been added to the quadrant multiplier of the P.F.C. controller U1A in order to provide a constant voltage control loop bandwidth of the P.F.C. controller U1A in spite of fluctuations that may be present on signal paths 18 and 20. A quadrant multiplier input allows external compensation to be applied to the current modulation quadrant. The oscillator within the P.F.C. controller U1A operates at a modulated switching frequency. Thus, the frequency of the PWM control signal output has its minimum, nominal value of approximately 20 KHz when the input voltage from rectifier V1 is at its minimum, zero value, and the frequency of the PWM control signal output has its maximum value of approximately 40 KHz when the input voltage from rectifier V1 is at its peak value.

[0041] The frequency of the pulse-width modulated control signal produced by the P.F.C. controller U1A, determines the current drawn from the full-wave rectifier V1 and hence from the AC supply line. By forcing the current produced by the P.F.C. controller U1A present on signal paths 24 and 26 to vary in accordance with the applied line voltage, the line current is forced to become sinusoidal, endowing the power factor correction circuit 22 with a near-unity power factor and low harmonic distortion. The output voltage value of the power factor correction circuit 22 can be adjusted by means of the resistor divider R9, R10, R8, and is typically set at 450V d.c. The output of the power factor correction circuit is routed, via signal paths 24 and 26 to the plurality of inverter power circuits 28A, 28B, 28C, and 28D, each comprised of a plurality of elements shown in Table 3 and each of which may be further described with reference to FIG. 3.

[0042]

	TABLE 3
	Element	Typical Value/Type
	C10	Film Capacitor
	C11	Film Capacitor
	C12	Film Capacitor
	C13	Film Capacitor
	C14	Film Capacitor
	C15	Film Capacitor
	C16	Electrolytic Capacitor
	C17	Film Capacitor
	C18	Film Capacitor
	C19	Film Capacitor
	C20	Film Capacitor
	R16	Resistor
	R17	Resistor
	R18	Resistor
	R19	Resistor
	R20	Resistor
	R21	Resistor
	R22	Resistor
	D4	Diode Fast Recovery
	Q4	Power Mosfet
	Q5	High Frequency
	L3	Power Mosfet
	U2A	Integrated Circuit Type IR21531D
	52	Fluorescent lamp

[0043] Each of the inverter power circuits 28A, 28B, 28C, and 28D comprises a half-bridge arrangement 46 made up of MOSFET devices Q4 and Q5, a half-bridge driver U2A, a resonant circuit 48 made up of capacitor C19 and inductor L3, all as shown in FIG. 3, as well as in FIG. 1.

[0044] The MOSFET devices Q4 and Q5 each have first (1) second (2) and third (3) electrodes. The MOSFET Q4 has its first (1) electrode connected to the positive d.c. voltage present on signal path 24, its third (3) electrode connected to the first end of the resonant circuit 48, whereas the first (1) electrode of the MOSFET Q5 is also connected to the first and of the resonant circuit 48, and the third (3) electrode of the MOSFET Q5 is connected to a ground connection (GND).

[0045] The half-bridge U2A provides first and second outputs that control the operation of a half-bridge arrangement 48, which operatively provides an oscillating current to energize the fluorescent lamp 52. The resonant circuit 48 has a second end connected to a first cathode of the fluorescent lamp 52, which, in turn, has a second cathode connected to the ground connection.

[0046] The half-bridge driver U2A has its pin 1 and pin 13 directly connected to a positive source of voltage 14V, to be later described with reference to FIG. 4, and serves as the controller for providing the circuit arrangement of FIG. 3 to serve as a self-oscillating half-wave.

[0047] Each of the inverter ballast circuits 28A, 28B, 28C, and 28D is set to oscillate at a fixed frequency of around 60 kHz determined by the net resistance-capacitance made up by the resistor R16 and the capacitor C13. In operation, when pin 1 and pin 13 of the half-wave driver U2A receives a positive voltage of +14V, and on the drain of the MOSFET Q4 receives a positive voltage of +450V, the inverter ballast circuits 28A, 28B, 28C, or 28D begins to oscillate and initially performs a frequency sweep to automatically look for the correct frequency of resonance (corresponding to that which energizes the fluorescent lamp into its running mode) determined by the values selected for the inductor L3, the capacitor C19 and the type of fluorescent lamp 52.

[0048] The U2A uses the following control sequence to drive rapid start fluorescent lamps. This sequence is composed from four steps: Startup, Preheat, Ignition Ramp and Run. During the startup mode, the operating frequency is determined by the parallel combination of Rph (R15), Rstart (R17) and Rt(R16), combined with the values of Cstart (C12), Ct(C13) and Rdt(R18). This frequency is normally chosen to ensure that the instantaneous voltage across the lamp during the first few cycles of operation does not exceed the strike potential of the lamp. During the preheat mode, the operating frequency is determined by the parallel combination of Rph (R15) and Rt(R16), combined with the value of Ct(C13) and Rdt (R18). This frequency, along with the preheat time, is normally chosen to ensure that adeguate heating of the lamp filament occur. At the end of the preheat time, the internal, open-drain transistor holding the Rph (R15) lead to ground turns off, and the voltage on this lead charges exponentially up to the Rt (Rt16) lead potential. During this ignition ramp mode, the output frequency is a function of the Rph (R15) and Cph (C10) time costant. The run mode operating frequency is therefore a function of the Ct(C13) and Rt (R16). Programming, opportunely, the sequence to drive of the IC U2A they are gotten two fundamentals benefits:

• [0049] 1) Auto adaptation of the circuit ballast. The ballast automatically looks for the correct frequency run so that to be able to operate any of the different types of fluorescent lamps, such as T5, T8, T12, 17W, 20W, 25W, 32W, 40W, 54W, 56W, 59W, 70W, linear, circular, 2D, twin tube or U-shaped fluorescent lamps.
• [0050] 2) Reducing the times of the sequence to drive is gotten a rapid ignition with possibility to operate the fluorescent lamp 52 by means of only two wires. In fact, from FIG. 3 it should be noted that the fluorescent lamp 52 is connected from one side to the 100 connection and from the other side to the fluorescent lamp 52 common ground. The half-bridge oscillator provided by each of the inverter ballast circuits 28A, 28B, 28C, and 28D feeds the fluorescent lamp in a series resonant arrangement, such particular configuration offer advantages of being adequately protected against possible damage owed to conditions of an absent lamp 52, broken cathodes in lamp 52, lighting failure of lamp 52, and interruption of connection of the lamp 52.

[0051] In the case wherein one of the above anomalies, for example, the half-bridge driver U2A finds itself without a load and instantly, in response to errors signals, performs a very rapid frequency sweep to find the new resonant condition, but being there is no load, the error signals related to the half-bridge driver U2A rise and accordingly the current produced by the half-bridge driver U2A for a brief time manifests excessive values, thereby jeopardizing the integrity of the inductor L3 and of the MOSFETS Q4 and Q5 that inevitably will destroy some component. This possible damage is prevented by the fault control logic contained in the IC U2A shown in FIG. 1, but more clearly in FIG. 3 as being comprised of components R22, R17, C14, R19 and C15 (56).

[0052] Three leads on the IC U2A are used for protection. These are SD (Unlatched Shutdown), CS (Latched Shutdown) and OC (Overcurrent). The first of these protection leads, SD, is used for both unlatched shutdown, this lead is connected to an external circuit which senses the presence of the lamp. When the SD lead exceeds the value (prefixed), signaling either a lamp fault or lamp removal, the oscillator is disabled, both gate driver outputs are pulled low, and the chip is put into the micropower mode.

[0053] The second lead used for protection is the CS lead, which is connected to a resistor Rcs (R22) in the source of the lower power MOSFET Q5. The CS lead is used to sense fault conditions such as failure of a lamp to strike, hard switching, no load, an operation below resonance. If any one of these conditions is sensed, the fault latch is set, the oscillator is disabled, the gate driver outputs so low, and the chip is put into the micro-power mode.

[0054] The third lead used for protection is the OC lead, used for the over-current, failure-to-strike, and hard switching fault condition. The level of this positive-going threshold is determined by the value the resistor Roc (R19).

[0055] To restore the condition of operation of the half-bridge driver U2A, it is only necessary to remove the cause that provoked the shutdown on half-bridge driver U2A, by for example, inserting a fluorescent lamp 52 or replacing a defective fluorescent lamp 52. The automatic reactivation of operation of the driver has not been inserted only for safety reasons intrinsic of the personnel employed to the maintenance. Obviously, the shutdown of the half-bridge driver U2A is separately activated for the inverter ballast circuits 28A, 28B, 28C and 28D. Accordingly, if an inverter ballast circuit, such as 28A, is deactivated, the remaining inverter ballast circuits 28B, 28C, and 28D remain fully operative.

[0056] It should now be appreciated that the practice of the present invention provides for an electronic ballast system that operates with the universal input having a voltage range of between 108 to 305 volts ac, and a frequency range of between 50–60 Hz. The electronic ballast system of the present invention allows for the rapid start of fluorescent lamps with preheating of the cathodes, thereby, allowing the electronic ballast to service the fluorescent lamps by the provisions of only two separate wires. Further, the electronic ballast system of the present invention having individual ballast circuits 28A, 28B, 28C, and 28D, allows for individual control of every fluorescent lamp whereas each of the individual ballast circuits 28A, 28B, 28C, and 28D provides for a selected power to a variety of fluorescent lamps, such as T5, T8, T12, 17W, 20W, 25W, 32W, 40W, 56W, 59W, 70W, linear, circular, 2D, twin tube or U-shape fluorescent lamps.

[0057] Between the PFC (FIG. 2) and the modules ballast 28A, 28B, 28C, 28D (FIG. 3) is inserted a static switch comprised of a plurality of elements shown in Table 4 and each of which may be further described with reference to FIG. 6.

[0058]

	TABLE 4
	R11	Resistor
	R12	Resistor
	R13	Resistor
	R14	Resistor
	R15	Resistor
	C10	Film Capacitor
	Q2	Transistor
	Q3	Transistor

[0059] The static switch has the purpose to send the signal of +14V, coming from the power supply (shown in FIG. 4) path 38, to the module ballast 28A, 28B, 28C, 28D paths 42 only when it is present the signal of +450V coming from the PFC line 24 (shown in FIG. 2) or the wall switch is turned on.

[0060] Through the sensor of high voltage, constituted by the pair resistors R11 and R12 the transistor Q2 is brought in conduction closing the circuit toward the ground sends in conduction the transistor Q3 through R13 and R14. Then the signal of 14V (Line 38) coming from the power supply passes through R15 (Line 42) and sent to the modules ballast. When the supply voltage is not present, to the line 24 are present a potential of 0V then the transistor Q2 and accordingly the transistor Q3 are interdicted not allowing send it some signal of 14V to the Vcc lead of the U2A present in the ballast modules 28A, 28B, 28C and 28D that accordingly they are out.

[0061] In the planning of the building, the integration of illumination for emergency purposes needs to be provided and such illumination must conform to various building codes and also to various legislative requirements. Particularly, these legislative requirements demand that when a building, that is frequented by the public, loses its primary source of power causing the loss of its ordinary illumination, then an auxiliary illumination is immediately provided for emergency purposes. Lighting for emergency purposes must clearly take into account the need to point out the routes of exit through customized signaling, the routes of emergency egress along hallways to allow a steady flow of people out of the exit of the related building, and to ensure that alarms and fire retarding equipment are clearly identified by the emergency lighting.

[0062] Typically, the emergency illumination is divided into safety illumination and backup components. The purpose of the safety light is to ensure the proper evacuation of the building and to assure that evacuation will be accomplished in a safe manner. The backup lighting commonly requires lighting to be provided on a continuous basis, so that various jobs may be performed by those employed in the related buildings.

[0063] Safety lighting is supplied, in the case of mains failure, within the extremely brief time of 0.5 seconds and this operation is accomplished automatically. So too is the return to normal power when the cause of failure has been corrected.

[0064] Commonly devices used in emergency lighting include nickel cadmium (Ni—Cd) that may be operated autonomously for at least an hour and have an operational life of at least four (4) years.

[0065] The electronic ballast system 10 of the present invention includes a switching power and battery charger 30 and an emergency section 32 that allows the fluorescent lamp of the emergency section to be energized by its battery. System 10, the emergency system of this invention, provides that when the main fails, one (1) fluorescent lamp will be kept alight by the automatic operation of the system. So that if the established configuration of the lighting has four (4) lamps, one (1) lamp will remain alight, if three (3) lamps, one (1) will remain on, if two (2) lamps, one (1) will take over. The emergency section 32 of the present invention provides power for fluorescent lamps, which can be of the 18 W to 70 W type. The emergency illumination of the present invention may be further described with reference to FIG. 4, showing the switching power and battery charger 30, which is comprised of a plurality of elements given in Table 5.

[0066]

	TABLE 5
	Element	Typical Value/Type
	C21	Electrolytic Capacitor
	C22	Electrolytic Capacitor
	C23	Film Capacitor
	C25	Film Capacitor
	C26	Electrolytic Capacitor
	R23	Resistor
	R24	Resistor
	D4	Diode Fast Recovery
	D5	Diode Fast Recovery
	D6	Diode Zener
	T1	High Frequency Transformer
	OC1	Optocoupler
	U3A	Integrated Circuit

[0067] FIG. 4 shows the switching power and battery charger 30 as having its inputs connected to the hot and ground terminals of the EMI filter 12 of FIG. 2. It is of particular importance that the hot connection of the switching power and battery charger 30 be maintained so as to not interrupt the function of the supplying emergency supply.

[0068] An essential element of the switching power and battery charger 30 is the fly-back controller U3A, which in an integrated circuit and operatively cooperates with other components shown in FIG. 4, as well as components shown in FIG. 5 to be described, to produce a stabilized dc voltage having a value of about +14 volts derived from an ac source having a voltage variation of between 108 to 305 volts ac, which is essentially the same ac source of voltage feeding the EMI filter 12. The fly-back controller U3A includes a high voltage (700V) power MOSFET switch with an internal power ballast controller. Unlike the P.F.C. controller U1A of FIG. 2, employing a conventional pulse-width modulator, the fly-back controller U3A uses a simple ON/OFF control to regulate the voltage output provided on its signal paths 34, 36 and 38. The fly-back controller U3A consists of an oscillator, an enable circuit, an under-voltage circuit, an over-temperature protection arrangement, a current limit circuit, and the 700 volt power MOSFET. The fly-back controller U3A is interconnected to the circuit elements of FIG. 4 by way of pins 1–4.

[0069] Pin 1 (D) of the fly-back controller U3A serves as its MOSFET drain connection and provides a signal for creating internal operating current for both start-up and steady-state operation. Pin 2 (EN/UV) of the fly-back controller U3A provides an under-voltage and enable function. The under-voltage internal circuitry disables the power MOSFET when the bypass pin (pin 3) voltage drops below a prefixed voltage value. The enable function of pin 2 determines whether or not to proceed with a next switch cycle. In operation, once a cycle is started the fly-back controller U3A always completes the cycle (even when enable pin 2 changes state halfway through the cycle). The enable signal of pin 2 is generated on the secondary of the transformer T2 (by way of pin 2 of the fly-back controller U3A) by comparing the power supply output voltage present on signal paths 34, 36 and 38 with an internal reference voltage. The enabler signal is high when the power supply output voltage is less than the reference voltage. This pin (2) is driven by an optocoupler (OC1) having a transistor. The collector of the optocoupler transistor is connected to the enable pin (2) and the emitter of the optocoupler transistor is connected to the source pin (5). The optocoupler has an LED which is connected in series with a zener diode (D6) across the DC output voltage present on signal paths 34, 36 and 38 when the DC output voltage exceeds a target or predetermined regulation voltage level (octocoupler LED diode voltage drop plus zener voltage). For such conditions the optocoupler LED will start to conduct, pulling the enable pin (1) low.

[0070] Pin 3 (BP) of the fly-back controller U3A serves as a BYPASS PIN, which is connected to a bypass capacitor C25. Pin 4 (S) of the fly-back controller serves as a SOURCE PIN for the internal power MOSFET.

[0071] The switching power and battery charger 30 operates from a full-wave rectifier V2 that receives the ac source (108–305 vac). Capacitor C21 and C22 filters the rectified dc output from V2 and provides a delaying function to compensate for delays in the standby power provided by the switching power and battery charger 30 involved in the loss of the primary power. The rectified d.c. voltage is also applied to the primary winding of transformer T1 arranged in series with an integrated high voltage MOSFET inside of the fly-back controller U3A. A diode D4, a capacitor C23 and a resistor R23 comprise a clamping circuit that limits the turnoff voltage spikes presented to the fly-back controller U3A (drain pin) to a safe value. The voltage at the secondary winding of transformer T1 is rectified and filtered by D5 and C26 to provide a 14.5 Vdc output. The output voltage is determined by the sum of the optocoupler OC1 LED forward drop voltage and zener diode D6 voltage drop. The resistor R24, maintains a bias current through the zener diode D6 so as to improve its voltage response.

[0072] The fly-back controller U3A is intended to operate in a current limit mode. When enabled, the oscillator within the fly-back controller U3A turns the internal power MOSFET on at the beginning of each cycle. The internal MOSFET is turned off when the current ramps up to a predetermined current limit. The maximum on-time of the MOSFET is limited to a DC maximum voltage associated with the oscillator. Since the current limit and frequency response of a given fly-back controller U3A are constant, the power delivered is proportional to the primary inductance of the transformer T1 and this power is relatively independent of the input voltage (108–305 ac.). This primary inductance of the transformer T1 is calculated for the maximum power required and is supplied to the oscillator within the fly-back controller U3A. As long as the fly-back controller U3A is chosen to be rated for the power level at the lowest input voltage, the primary inductance will ramp up the current to the current limit before the voltage DC maximum limit of the associated oscillator is reached. The emergency section 32 of the electronic ballast circuit 10, may be further described with reference to FIG. 5 showing a plurality of elements given in Table 5.

[0073]

	TABLE 6
	Element	Typical Value/Type
	R27	Resistor
	R28	Resistor
	R29	Resistor
	R30	Resistor
	C27	Electrolytic Capacitor
	C28	Film Capacitator
	C29	Film Capacitator
	C30	Film Capacitor
	D8	Diode Rectifier
	D7	Diode LED
	K1	Relay Double-Pole, Double-Throw
	T2	High Frequency Transformer
	Q6	Power Mosfet
	Q7	Power Mosfet
	LP1	Fluorescent Lamp
	Battery	Nickel-cadmium Type (12 volts)

[0074] In general, an essential feature of the emergency section 32 is to act as a high frequency inverter that converts the dc output of the battery into an alternating voltage having a typical value of 190 volts and a typical frequency of 50 KHz having a relatively low value of current. The high frequency inverter is used to turn on the fluorescent lamp LP1 being of a typical low-power (8 W) type. The enablement of the usage of the battery is determined by the commutation provided by the switch control K1. The energizing of the switch control K1 causes the normally open contact SC1 to interrupt the positive voltage battery 110 with the bus positive of the emergency circuit 140. The normally closed contact SC1 related to the switch control K1 connects to the node 110 to node 140. Further, the switch control K1 as shown in FIG. 1 has a second set of contacts, in particular normally closed contact SC2, which connects junction 200 to the lamp LP1, as well as to the fluorescent lamp 52 of FIG. 3 by way of signal path 44. The high frequency inverter of the emergency section 32 utilizes a half-bridge driver U4 whose operation and internal components are the same as previously discussed for the controller U2A (see circuit 28A, 28B, 28C, 28D). The half-bridge driver U4A is operatively connected to MOSFETS Q6 and Q7 arranged in a half-bridge configuration 40 controlled, in part, by a capacitator C29 that is interconneted to the primary of the transformer T2 by way of capacitor C30. The capacitance of the capacitor C30 and the inductance of transformer T2 comprise a resonant circuit. The half-bridge driver U4A is interconnected to the circuit of FIG. 5 by way of its pins 1–8.

[0075] Pin 1 (Vcc) of the half-bridge driver U4A serves as the supply source for logic and a gate driver all within the half-bridge driver U4A. This pin 1 is connected to the positive node 150 and to the capacitor C27. Pin 2 (RT) of the half-bridge driver U4A serves as an oscillator timing resistor input. This pin 2 is connected to the resistor R28. Pin 3 (CT) of the half-bridge driver U4A serves as an oscillator timing capacitor input. This pin 3 is connected to capacitor C28. Pin 4 (COM) of the half-bridge driver U4A serves as an IC power and signal ground. This pin 4 is directly connected to the ground common of the circuit of FIG. 5. Pin 5 (LO) of the half-bridge driver U4A serves as a low side to gate the output which drives MOSFET Q7. This pin 5 is connected to the gate of the MOSFET Q11 by resistor R30.

[0076] Pin 6 (VS) of the half-bridge driver serves a high voltage floating supply return. This pin 6 is connected to the junction between the drain of the MOSFET Q6 and the source of the MOSFET Q7, as well as to capacitor C29. Pin 7 (HO) of the half-bridge driver U4A serves as a high side to gate the MOSFET Q6. This pin 7 is connected to the gate of the MOSFET Q6 by resistor R29. Pin 8 (VB) of the half-bridge driver U4A serves as a high side to gate the MOSFET Q7 with a floating supply. This pin 8 is also connected to the capacitor C29.

[0077] In operation, that is, in case of a primary power failure, when the half-bridge driver U4A receives a positive voltage of about 13.6 volts coming from a battery, it begins to alternately oscillate at a fixed frequency of about 50 kHz determined by the value of capacitor C28 and resistor R28. This oscillation drives the two MOSFETS Q6 and Q7 arranged in a half-bridge configuration 40. The application of the voltage of 13.5 applied to the drain of Q6 (positive) and to the source of Q7 (negative) causes the half-bridge driver U4A to convert an alternating voltage to a square type form having the high frequency 50 kHz. The high frequency is established by the resonant circuit formed by the capacitor C30 and the primary inductance of transformer T2. The capacitor C30 serves to limit the current flowing in the transformer T2. The high frequency voltage present at the capacitor C25 is transferred to the secondary of transformer T2 by an appropriate selection of the ratio of the windings between the primary and secondary of transformer T2. This voltage appearing across the secondary winding of transformer T2 is raised to a value sufficient to render the fluorescent lamp LP1 into its operating state again without first sequencing the fluorescent through its preheat mode.

[0078] The emergency section 32 of FIG. 3 allows the fluorescent lamp LP1 be placed into its running mode and eliminates any annoying negative effects created by flickering typically experienced in low power fluorescent lamps, such as LP1. Conversely, the operation of the present invention causes the fluorescent lamp LP1, to provide a constant homogenous brightness.

[0079] The battery, shown in FIG. 5, has positive and negative terminals and may be of a nickel-cadmium type having twelve cells, each being 1.2 volts connected in series so as to provide a total current capability of 3 amps-hours at 12 volts dc. The typical consumption of current by the circuit arrangement of FIG. 5 varies from a minimum of 0.5 amps (a lamp LP1 of 18 W) to a maximum of 0.9 amps (a lamp LP1 of 60 W). The battery for such variations can operate autonomously for at least one hour.

[0080] The current to charge the battery is approximately and preferably regulated to about ⅛ of an amp and provides an output of approximately 13.6 volts in response to a trickle charge of about twelve (12) hours. It is possible to increase the autonomy of the battery by selecting a battery having a maximum of seven (7) amp hours and allowing for a charge to last for about four (4) hours.

[0081] In the overall operation of the present invention, when the primary power source is available, the switching power and battery charger 30 of FIG. 4 provides an input of 14 volts to the emergency section shown in FIG. 5. This input causes the indicator (ON) of the status of the battery charger to be display by LED D7 and the resistor R27. The presence of this 14 volts excites the relay K1 thereby commutating the contacts SC1 and SC2 causing normally opened contact SC1. Conversely, the presence of the energized coil of switch control K1 causes the normally closed contact to open up, thereby removing the 14 volts to the half-bridge driver U4A. Further, the energizing of the switch control K1 causes the normally closed contact SC2 to open up, thereby removing the power from node 200 to the lamp LP1 and also to the lamp 52 of FIG. 3, otherwise energized by way of signal path 44.

[0082] Conversely, when the primary supply voltage is absent, the switch power and battery charger 30 does not produce the 14 volts otherwise applied to the emergency section 32 shown in FIG. 5. This absence causes the LED D7 to extinguish. Further, the absence of the 14 volts from the switch power and battery charger 30 removes the excitation to the coil of relay K1, thereby allowing its normally closed contact SC1 and SC2 to be functional. The normally closed contact SC1, as shown in FIG. 5, causes the voltage from the battery to be delivered to node 140, which, in turn, is passed to the half-bridge driver U4A causing it to operate and supplying a oscillating current present on node 200 to the fluorescent lamp LP1 rendering it into its operating condition. Further, the oscillating current present on node 200 is delivered, via signal path 44 to lamp 52 of the selected inverter ballast circuit 28A, 28B, 28C, or 28D.

[0083] The transition of the emergency section 32 of FIG. 5 from its dormant condition, that is, the primary power being present, to its active condition, that is, when the primary power is lost, is dependent upon commutation time, which is typically 20 milliseconds for the relay K1. During the operation that provides for the emergency lighting, to avoid the battery from completely discharging so as to provide for a prolonged blackout, the emergency section of FIG. 5 utilizes a battery sensor circuit integrated in the half-bridge driver U4A. More particularly, pin 1 of the half-bridge driver U4A is internally connected and constantly monitors the d.c. voltage feeding the half-bridge driver U4A, and when this d.c. voltage drops below a particular value, such as 8.2 volts, the battery sensor immediately extinguishes the oscillator within the half-bridge driver U4A, thereby, setting the half-bridge driver U4A into its standby condition not energizing the fluorescent lamp LP1. The functionality or operational condition of the half-bridge driver U4A is restored when the voltage present on pin 1 again rises to a value of approximately 11 volts and then the battery begins its charge. Such a mechanism provides for the advantages in that it avoids damage to the cells of the battery, otherwise created by a complete discharge thereof, increases in the time for recharge, decreases in the value of current necessary for the recharge of the battery and correspondingly increases the operation life of the battery.

[0084] It should now be appreciated that the practice of the present invention provides for an electronic ballast system 10 having a subsystem with a battery, along with a battery charger and associated logic, that provides for emergency lighting, as well as for operating at least one of the fluorescent lamps of the inverter ballast circuits 28A, 28B, 28C, and 28D.

[0085] It is understood that the invention is not limited to the specific embodiments herein illustrated and described, and that some changes and modifications may be suggested which are obvious from the description herein. Such changes and modifications are nevertheless within the scope of the present invention.