HAVING INTEGRAL ELECTRONIC BALLAST CIRCUIT

BACKGROUND OF THE INVENTION Cross References to Related Applications

[0001] This application is a non-provisional application of commonly-assigned U. S.

Provisional Patent Application No. 61/484,481, filed May 10, 2011, entitled DIMMABLE SCREW-IN COMPACT FLUORESCENT LAMP HAVING INTEGRAL ELECTRONIC BALLAST CIRCUIT, the entire disclosures of which are hereby incorporated by reference.

Field of the Invention

[0002] The present invention relates to electronic ballasts for gas discharge lamps, such as fluorescent lamps, and particularly, a dimmable screw-in compact fluorescent lamp having an integral electronic ballast circuit.

Description of the Related Art

[0003] In order to save energy, high-efficiency lighting loads, such as, for example, compact fluorescent lamps (CFLs) and light-emitting diode (LED) light sources, are being used in place of or as replacements for conventional incandescent or halogen lamps. High-efficiency light sources typically consume less power and provide longer operational lives as compared to incandescent and halogen lamps. Fig. 1 is a simplified block diagram of a prior art lighting control system 10 having a screw-in compact fluorescent lamp 20. The screw-in compact fluorescent lamp 20 comprises a fluorescent lamp tube 22, which may be formed in a spiral (as shown in Fig 1). The screw-in compact fluorescent lamp 20 also comprises an enclosure 24 for housing a load regulation circuit 40 (Fig. 2), e.g., an electronic ballast circuit, which is electrically coupled to the lamp tube 22 for illuminating the lamp tube. The screw-in compact fluorescent lamp 20 has a screw-in base 26 adapted to be coupled to a standard Edison socket. The lamp tube 22 of a typical prior art screw-in compact fluorescent lamp 20 is filled with a fill gas of 100% argon at a pressure of approximately 4 Torr. [0004] The lighting control system 10 also comprises a "two-wire" dimmer switch 30, which is coupled in series between an alternating-current (AC) power source 15 and the screw-in compact fluorescent lamp 20 for controlling the intensity of the lamp tube 22. The dimmer switch 30 may be adapted to be mounted to a standard electrical wallbox and comprises a hot terminal H coupled to the AC power source 15 for receiving an AC mains line voltage VAC, and a dimmed-hot terminal DH coupled to the screw-in compact fluorescent lamp 20. The dimmer switch 30 does not require a direct connection to the neutral side N of the AC power source 15. Examples of prior art dimmer switches are described in greater detail is commonly- assigned U.S. Patent No. 5,248,919, issued September 29, 1993, entitled LIGHTING CONTROL DEVICE; U. S. Patent No. 6,969,959, issued November 29, 2005, entitled ELECTRONIC CONTROL SYSTEMS AND METHODS; and U.S. Patent No. 7,687,940, issued March 30, 2010, entitled DIMMER SWITCH FOR USE WITH LIGHTING CIRCUITS HAVING THREE-WAY

SWITCHES, the entire disclosures of which are hereby incorporated by reference.

[0005] The dimmer switch 30 comprises a bidirectional semiconductor switch 32 coupled between the hot terminal H and the dimmed hot terminal DH for conducting a load current ILOAD through the screw-in compact fluorescent lamp 20. The bidirectional semiconductor switch 32 may comprise a single device, such as a triac, or a combination of devices, such as, two field- effect transistors (FETs) or insulated-gate bipolar junction transistors (IGBTs) coupled in anti- series connection. To control the amount of power delivered to the lamp tube 22, the bidirectional semiconductor switch 32 is controlled to be conductive and non-conductive for portions of a half-cycle of the AC power source 15, such that the bidirectional semiconductor switch is rendered conductive for a conduction time TON each half-cycle. The dimmer switch 30 may comprise a toggle actuator for turning the high-efficiency lamp tube 22 on and off and an intensity adjustment actuator for adjusting the intensity of the lamp tube 22 between a low-end intensity and a high-end intensity.

[0006] The dimmer switch 30 further comprises a control circuit 34 coupled in parallel with the bidirectional semiconductor switch 32 for conducting a control circuit ICNTL through the screw-in compact fluorescent lamp 20. The control circuit 34 is coupled to a control input of the bidirectional semiconductor switch 32 for rendering the bidirectional semiconductor switch conductive and non-conductive to generate a phase-control voltage Vpc using either the forward phase-control dimming technique or the reverse phase-control technique. Accordingly, the bidirectional semiconductor switch 32 is rendered conductive for the conduction time T _C ON each half-cycle, thus setting a duty cycle DCpc of the phase-control voltage Vpc. The control circuit 34 may be operable to provide, for example, a constant gate drive to the bidirectional semiconductor switch 32, such that the bidirectional semiconductor switch will remain conductive independent of the magnitude of the load current ILOAD conducted through the bidirectional semiconductor switch and the screw-in compact fluorescent lamp 20. An example of a two-wire dimmer switch having a constant gate drive control circuit is described in greater detail in commonly-assigned, co-pending U.S. Patent Application No. 12/952,920, filed November 23, 2010, entitled TWO-WIRE DIMMER SWITCH FOR LOW-POWER LOADS, the entire disclosure of which is hereby incorporated by reference.

[0007] The screw-in base 26 of the compact fluorescent lamp 20 provides for only two electrical connections: a phase-control connection PC to the dimmer switch 30 for receiving the phase-control voltage Vpc and a neutral connection NC to the neutral side N of the AC power source 15. The load regulation circuit 40 of the screw-in compact fluorescent lamp 20 is operable to adjust the intensity of the lamp tube 22 between the low-end intensity and the high- end intensity in response to the duty-cycle DCpc of the phase-control signal Vpc (i.e., the conduction time of the bidirectional semiconductor switch 32 of the dimmer switch 30).

[0008] With forward phase-control dimming, the bidirectional semiconductor switch 32 is rendered conductive at some point within each AC line voltage half-cycle and remains conductive until approximately the next voltage zero-crossing, such that the bidirectional semiconductor switch is conductive for the conduction time each half-cycle. A zero-crossing is defined as the time at which the AC line voltage transitions from positive to negative polarity, or from negative to positive polarity, at the beginning of each half-cycle. Forward phase-control dimming is often used to control energy delivered to a resistive or inductive load, which may include, for example, an incandescent lamp or a magnetic low-voltage transformer. The bidirectional semiconductor switch of a forward phase-control dimmer switch is typically implemented as a thyristor, such as a triac or two silicon-controlled rectifiers (SCRs) coupled in anti-parallel connection, since a thyristor becomes non-conductive when the magnitude of the current conducted through the thyristor decreases to approximately zero amps. Thyristors are typically characterized by a rated latching current and a rated holding current, and comprise two main terminals and a control terminal. The current conducted through the main terminals of the thyristor must exceed the latching current for the thyristor to become fully conductive. In addition, the magnitude of the load current ILOAD conducted through the main terminals of the thyristor must remain above the holding current for the thyristor to remain in full conduction.

[0009] The control circuits of many forward phase-control dimmers comprise analog control circuits (such as timing circuits) for controlling when the thyristor is rendered conductive each half-cycle of the AC power source. The analog control circuit typically comprises a potentiometer, which may be adjusted in response to a user input provided from, for example, a linear slider control or a rotary knob in order to control the amount of power delivered to the lighting load. The analog control circuit is typically coupled in parallel with the thyristor and conducts a small timing current through the lighting load when the thyristor is non-conductive.

[0010] With reverse phase-control dimming, the bidirectional semiconductor switch 32 is rendered conductive at the zero-crossing of the AC line voltage and rendered non-conductive at some point within each half-cycle of the AC line voltage, such that the bidirectional

semiconductor switch is conductive for a conduction time each half-cycle. The bidirectional semiconductor switch of reverse phase-control dimmers typically comprises two field-effect transistors (FETs) in anti-serial connection, or the like. Accordingly, prior art reverse phase- control dimmer switches have required advanced control circuits (such as microprocessors) for controlling the operation of the FETs, and power supplies for powering the microprocessors. In order to properly charge, the power supply of such a two-wire dimmer switch must develop an amount of voltage across the power supply and must conduct a charging current from the AC power source through the electrical load, in many instances even when the lighting load is off.

[0011] Fig. 2 is a simplified block diagram of the load regulation circuit 40 of the prior art screw-in compact fluorescent lamp 20. The load regulation circuit 40 comprises an electromagnetic interference (EMI) filter 50 for preventing noise generated by the load regulation circuit from being conducted on the AC mains wiring. A full-wave bridge rectifier 52 receives the phase-control voltage Vpc from the EMI filter 50 and generates a rectified voltage VRECT- The rectified voltage VRECT is coupled to a bus capacitor CBUS through a diode D54 for generating a direct-current (DC) bus voltage VBUS across the bus capacitor. The load regulation circuit 40 further comprises an inverter circuit 56 for generating a high-frequency square-wave voltage VSQ from the rectified voltage VRECT, and a resonant tank circuit 58 for receiving the square-wave voltage VSQ and producing a substantially sinusoidal high-frequency AC voltage VLAMP (i.e., an arc voltage or lamp voltage), which is provided to the lamp tube 22. The inverter circuit 56 adjusts the operating frequency fop of the square-wave voltage VSQ in order to adjust the intensity of the lamp tube 22.

[0012] The load regulation circuit 40 further comprises a phase-to-DC converter circuit

60 for receiving the rectified voltage VRECT and generating a DC voltage VDC that has a magnitude that is representative of the duty-cycle DCpc of the phase-control signal Vpc, and a lamp current sense circuit 62 that generates a lamp current control signal VILAMP representative of a magnitude of a lamp current ILAMP conducted through the lamp tube 22. A control circuit 64 is coupled to the inverter circuit 56 for adjusting an operating frequency fop of the square wave voltage VSQ and thus the magnitude of the lamp current ILAMP in response to the duty-cycle DCpc of the phase-control signal Vpc and the magnitude of the lamp current ILAMP- The load regulation circuit 40 also comprises a power supply 66 that receives the bus voltage VBUS and generates a DC voltage Vcc for powering the control circuit 64.

[0013] Since the dimmer switch 30 is a two-wire dimmer switch, the compact fluorescent lamp 20 receives both power for energizing the lamp tube 22 and information for determining the target intensity of the lamp tube from the phase-control signal Vpc. The phase-to-DC circuit 60 typically comprises a filter circuit for preventing voltage fluctuations in the AC mains line voltage VAC of the AC power source 15 or noise on the AC mains line voltage VAC from altering the magnitude of the DC voltage VDC generated by the phase-to-DC converter circuit 60.

Therefore, there is typically a delay time period between a change in the duty-cycle DCpc of the phase-control signal Vpc and a resulting change in the magnitude of the DC voltage VDC

generated by the phase-to-DC converter circuit 60. If the intensity adjustment actuator of the dimmer switch 30 is controlled such that the target intensity is quickly reduced from the high-end intensity to the low-end intensity, the magnitude of the phase-control signal Vpc (and thus the amount of power available to the load regulation circuit 40) will quickly decrease while the control circuit 64 is still controlling the intensity of the lamp tube 22 to the high-end intensity (due to the delay time period). Accordingly, the bus capacitor CBUS will quickly discharge, such that the control circuit 64 becomes unpowered and the lamp tube 22 is extinguished, which, of course, is undesirable.

[0014] Figs. 3A and 3B show example waveforms of the DC bus voltage VBUS and the lamp current ILAMP, respectively. As shown in Fig. 3A, the bus voltage VBUS is characterized by some low-frequency voltage ripple having a frequency approximately equal to twice the frequency of the AC power source 15 (e.g., approximately 120 Hz). The control circuit 64 is typically characterized by a corner frequency of approximately 10-20 Hz, and thus controls the inverter circuit 56 to adjust the operating frequency fop of the square-wave voltage VSQ at a relatively slow rate in response to the lamp current control signal VILAMP- Since the corner frequency (i.e., approximately 10-20 Hz) is less than the frequency of the voltage ripple of the bus voltage VBUS (i.e., approximately 120 Hz), the operating frequency fop of the square-wave voltage VSQ (and thus the lamp current ILAMP) is maintained relatively constant over short time intervals (e.g., during a single half-cycle T _H c of the AC power source 15, i.e., approximately 8.33 msec). As a result, the lamp current ILAMP has an envelope IENV that is characterized by the frequency of the voltage ripple of the bus voltage VBUS (as shown in Fig. 3B). This fluctuation (or ripple) in the envelope IENV of the lamp current ILAMP can cause undesirable flicker in the lamp tube 22.

[0015] Accordingly, there is a need for a dimmable screw-in compact fluorescent lamp having an integral electronic ballast circuit that avoids the disadvantages of the prior art circuits.

SUMMARY OF THE INVENTION

[0016] According to an embodiment of the present invention, a dimmable ballast circuit for a compact fluorescent lamp generates a lamp current through the lamp having a substantially constant envelope such that flicker in the lamp electromagnetic interference (EMI) noise on an AC voltage supply is minimized. The ballast circuit comprises a bus-voltage-generating circuit receiving an input from an AC voltage supply and producing a DC bus voltage characterized by a low-frequency voltage ripple, an inverter circuit receiving the DC bus voltage and generating a variable frequency square-wave output signal, and an output filter circuit receiving the square- wave output signal and generating a substantially sinusoidal lamp current through the lamp at the frequency. The inverter circuit has a control input provided with a drive control signal that determines the frequency of operation of the inverter circuit and thus the frequency of the square- wave output signal. The ballast circuit further comprises a control circuit producing the drive control signal provided at the control input of the inverter circuit to control the frequency of the square-wave output signal and thus the intensity level of the lamp. The control circuit receives a first signal related to a target intensity level of the lamp and a second signal related to the actual intensity level of the lamp. The control circuit comprises an error amplifier circuit for comparing the first and second signals to produce the drive control signal to cause the inverter circuit to operate at a frequency to illuminate the lamp at the target intensity level. The bus-voltage- generating circuit produces the DC bus voltage to define an envelope for the sinusoidal lamp current, and the error amplifier circuit operates in a frequency range such that the error amplifier circuit is responsive to the second signal at the frequency of the second signal thereby to adjust the frequency of operation of the inverter circuit to reduce variation in the envelope of the sinusoidal lamp current.

[0017] In addition, a dimmable compact fluorescent lamp adapted to receive a phase- control voltage from a dimmer switch is also described herein. The dimmable compact fluorescent lamp comprises a fluorescent lamp tube, and an electronic ballast circuit for receiving the phase-control voltage from the dimmer switch and illuminating the lamp tube in response to the phase-control voltage. The ballast circuit comprises an inverter circuit for generating a variable frequency square-wave output signal from a DC bus voltage and an output filter circuit receiving the square wave output signal and generating a substantially sinusoidal lamp current through the lamp tube at the frequency. The electronic ballast circuit further comprises a control circuit for producing a drive control signal that is provided to a control input of the inverter circuit to control the frequency of the square-wave output signal and thus the intensity level of the lamp tube. The control circuit receives a first signal related to a target intensity level of the lamp tube and a second signal related to the actual intensity level of the lamp tube. The control circuit comprises an error amplifier circuit for comparing the first and second signals to produce the drive control signal to cause the inverter circuit to operate at a frequency to illuminate the lamp tube at the target intensity level. The DC bus voltage defines an envelope for the sinusoidal lamp current, and the error amplifier circuit operates in a frequency range such that the error amplifier circuit is responsive to the second signal at the frequency of the second signal thereby to adjust the frequency of operation of the inverter circuit to reduce variation in the envelope of the sinusoidal lamp current.

[0018] According to another embodiment of the present invention, a lighting control system receiving power from an AC power source comprises a dimmable compact fluorescent lamp and a dimmer switch adapted to be coupled in series electrical connection between the AC power source and the dimmable compact fluorescent lamp. The dimmable compact fluorescent lamp includes a fluorescent lamp tube and an electronic ballast circuit for illuminating the lamp tube. The dimmer switch generates a phase-control voltage characterized by a duty cycle defining a target intensity level of the lamp tube of the dimmable compact fluorescent lamp. The electronic ballast circuit receives the phase-control voltage and illuminates the lamp tube in response to the phase-control voltage. The ballast circuit comprises an inverter circuit for generating a variable frequency square-wave output signal from a DC bus voltage and an output filter circuit receiving the square-wave output signal and providing a substantially sinusoidal lamp current through the lamp tube at the frequency. The electronic ballast circuit further comprises a control circuit for producing a drive control signal that is provided to a control input of the inverter circuit to control the frequency of the square-wave output signal and thus the intensity level of the lamp tube. The DC bus voltage defines an envelope for the sinusoidal lamp current, and the control circuit operates in a frequency range such that the control circuit adjusts the frequency of operation of the inverter circuit to reduce variation in the envelope of the sinusoidal lamp current.

[0019] According to another aspect of the present invention, a dimmable ballast circuit for a fluorescent lamp comprises: (1) an inverter circuit generating a variable frequency square- wave output signal, the inverter circuit having a control input provided with a drive control signal that determines the frequency of operation of the inverter circuit and thus the frequency of the square -wave output signal; (2) an output filter circuit receiving the square wave output signal and generating a substantially sinusoidal lamp current through the lamp at the frequency; (3) a non-linear amplifier circuit amplifying a lamp-current-feedback signal representative of the magnitude of the lamp current through the lamp; and (4) a control circuit producing the drive control signal provided at the control input of the inverter circuit to control the frequency of the square-wave output signal and thus the intensity level of the lamp. The control circuit receives a target-intensity signal related to a target intensity level of the lamp and compares the target- intensity signal and the lamp-current-feedback signal to produce the drive control signal. The control circuit operates at a bandwidth greater than approximately 10 kHz to cause the inverter circuit to operate at a frequency in order to illuminate the lamp at the target intensity level. For first values of the lamp current that are below a current threshold, the lamp-current-feedback signal is amplified by a first amount, and for second values of the lamp current that are above the current threshold, the lamp-current-feedback signal is amplified by a second, smaller amount, thereby increasing responsiveness of the control circuit to changes in lamp intensity levels at low lamp intensity levels. [0020] According to another embodiment of the present invention, a dimmable compact fluorescent lamp adapted to receive a phase-control voltage from a dimmer switch comprises a fluorescent lamp tube filled with a fill-gas mixture having a pressure of approximately 2 Torr and a fill-gas ratio of approximately 85 : 15 argon to neon, and an electronic ballast circuit having a control circuit that operates at a bandwidth greater than approximately 10 kHz. The electronic ballast circuit receives said phase-control voltage from said dimmer switch and illuminates said lamp tube in response to said phase-control voltage. The ballast circuit comprises an inverter circuit for generating a variable frequency square-wave output signal and an output filter circuit for receiving said square wave output signal and generating a substantially sinusoidal lamp current through said lamp tube at said frequency. The electronic ballast circuit further comprises a control circuit for producing a drive control signal that is provided to a control input of said inverter circuit to control the frequency of said square-wave output signal and thus the intensity level of said lamp tube. The control circuit receives a first signal related to a target intensity level of said lamp tube and a second signal related to the actual intensity level of said lamp tube, and compares said first and second signals to produce said drive control signal to cause said inverter circuit to operate at a frequency in order to illuminate said lamp tube at said target intensity level.

[0021] Other features and advantages of the present invention will become apparent from the following description of the invention that refers to the accompanying drawings.

BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The invention will now be described in greater detail in the following detailed description with reference to the drawings in which:

[0023] Fig. 1 is a simplified block diagram of a prior art lighting control system including a "two-wire" dimmer switch for controlling the amount of power delivered to a screw-in compact fluorescent lamp;

[0024] Fig. 2 is a simplified block diagram of the screw-in compact fluorescent lamp of the lighting control system of Fig. 1 ;

[0025] Fig. 3A shows an example waveform of a bus voltage of the screw-in compact fluorescent lamp of Fig. 2; [0026] Fig. 3B shows an example waveform of a lamp current of the screw-in compact fluorescent lamp of Fig. 2;

[0027] Fig. 4 is a side view of a dimmable screw-in compact fluorescent lamp according to an embodiment of the present invention;

[0028] Fig. 5 is a simplified schematic diagram of the screw-in compact fluorescent lamp of Fig. 4;

[0029] Fig. 6 is a simplified schematic diagram of the electrical circuitry of the screw-in compact fluorescent lamp of Fig. 4 showing a load regulation circuit and two power supplies in greater detail;

[0030] Fig. 7 is a simplified schematic diagram of an inverter circuit of the load regulation circuit and an over-voltage protection circuit of the screw-in compact fluorescent lamp of Fig. 4;

[0031] Fig. 8 is a simplified schematic diagram of a control circuit of the screw-in compact fluorescent lamp of Fig. 4;

[0032] Fig. 9 is an example plot of the magnitude of a lamp voltage with respect to the magnitude of a lamp current of the screw-in compact fluorescent lamp of Fig. 4;

[0033] Fig. 10A shows an example waveform of a bus voltage of the screw- in compact fluorescent lamp of Fig. 4; and

[0034] Fig. 10B shows an example waveform of a lamp current of the screw-in compact fluorescent lamp of Fig. 4 according to the embodiment of the present invention.

DETAILED DESCRIPTION OF THE INVENTION

[0035] The foregoing summary, as well as the following detailed description of the preferred embodiments, is better understood when read in conjunction with the appended drawings. For the purposes of illustrating the invention, there is shown in the drawings an embodiment that is presently preferred, in which like numerals represent similar parts throughout the several views of the drawings, it being understood, however, that the invention is not limited to the specific methods and instrumentalities disclosed. [0036] Fig. 4 is a side view of a dimmable screw-in compact fluorescent lamp 120 according to an embodiment of the present invention. The dimmable screw-in compact fluorescent lamp 120 comprises a light source, e.g., a fluorescent lamp tube 122, which may be formed in a spiral (as shown in Fig 4), in two or more U-bends, or in any other suitable form. The lamp tube 122 may be filled with a fill-gas mixture having a fall-gas pressure of

approximately 2 Torr and a fill-gas ratio of approximately 85: 15 argon to neon. The dimmable screw-in compact fluorescent lamp 120 further comprises an enclosure 24 for housing a load regulation circuit 130 (Fig. 5), e.g., an electronic ballast circuit, which is electrically coupled to the lamp tube 122 for illuminating the lamp tube.

[0037] The screw-in compact fluorescent lamp 120 has a screw-in base 126 adapted to be coupled to a standard Edison socket, such that the lamp is adapted to be coupled to a two-wire dimmer switch (such as the dimmer switch 30) via the phase-control connection PC of the screw- in base 126 and to the neutral side N of an AC power source via the neutral connection NC (as in the prior art lighting control system 100 shown in Fig. 1). As defined herein, a "two-wire" dimmer switch or load control device does not require a direct connection to the neutral side N of the AC power source. In other words, all currents conducted by the two-wire dimmer switch must also be conducted through the load. A two-wire dimmer switch may have only two terminals (i.e., the hot terminal H and the dimmed hot terminal DH as shown in Fig. 1).

Alternatively, a two-wire dimmer switch could comprise a three-way dimmer switch that may be used in a three-way lighting system and has at least three load terminals, but does not require a neutral connection. In addition, a two-wire dimmer switch may comprise an additional connection that provides for communication with a remote control device (for remotely controlling the dimmer switch), but does not require the dimmer switch to be directly connected to neutral.

[0038] Fig. 5 is a simplified schematic diagram of a dimmable screw-in compact fluorescent lamp 120 according to the embodiment of the present invention. As previously mentioned, the screw-in base 126 provides for only two electrical connections: the phase-control connection PC to the dimmer switch for receiving the phase-control voltage Vpc and the neutral connection NC to the neutral side N of the AC power source. The load regulation circuit 130 of the screw- in compact fluorescent lamp 120 is operable to adjust the intensity of the lamp tube 122 to a target intensity L _T RGT (i.e., a desired intensity) in response to the duty-cycle DCpc of the phase-control signal Vpc. The target intensity L _T RGT may range between a low-end intensity L _L E (e.g., approximately 1%) and a high-end intensity LRE (e.g., approximately 100%).

[0039] The screw-in compact fluorescent lamp 120 comprises a filter network 200 coupled to the phase-control connection PC and the neutral connection NC of the connector 126 for receiving the phase-control voltage Vpc from the dimmer switch 30. The filter network 200 comprises an inductor L201 (e.g., having an inductance of approximately 680 uH) and two capacitors C202, C203 (e.g., having capacitances of approximately 33 nF). The filter network 200 operates to prevent noise generated by the load regulation circuit 130 from being conducted on the AC mains wiring. The filter network 200 couples the phase-control voltage VPC to a voltage doubler circuit 205 (i.e., a bus-voltage-generating circuit), which generates a direct-current (DC) bus voltage VBUS across two series connected bus capacitors CBI, CB2. The first bus capacitor CBI conducts the load current ILOAD through a diode D206 (and the dimmer switch 30) to charge during the positive half-cycles, while the second bus capacitor CB ₂ conducts the load current ILOAD through a diode D208 to charge during the negative half-cycles.

Accordingly, the peak magnitude of the bus voltage VBUS is approximately twice the peak voltage of the AC mains line voltage VAC A half-bus voltage VHB is generated across the first bus capacitor CBI and has a magnitude equal to approximately half of the bus voltage VBUS-

[0040] The load regulation circuit 130 (i.e., the electronic ballast circuit) includes a half-bridge inverter circuit 210 for converting the DC bus voltage VBUS to a high-frequency square-wave voltage VSQ having an operating frequency fop. The load regulation circuit 130 further comprises an output filter circuit, e.g., a resonant tank circuit 220, for filtering the square- wave voltage VSQ to produce a substantially sinusoidal high-frequency AC voltage, which is coupled to the electrodes of the lamp tube 122. A control circuit 230 is coupled to the inverter circuit 210 for providing a drive control signal VDR to the inverter circuit 210 for adjusting the operating frequency fop of the square wave voltage VSQ and thus the magnitude of a lamp current ILAMP conducted through the lamp tube 122 in order to turn the lamp tube on and off and adjust the intensity of the lamp tube.

[0041] The screw-in compact fluorescent lamp 120 further comprises two power supplies: an inverter power supply 240 and a control power supply 250. The inverter power supply 240 receives the half-bus voltage VHB across the first bus capacitor CBI and generates a direct-current (DC) inverter supply voltage VL V (e.g., approximately 15 volts) for powering the control circuitry of the inverter circuit 210. The control power supply 250 draws current from the resonant tank circuit 220 and generates a DC control supply voltage Vcc (e.g., approximately 12 volts) for powering the control circuit 230 while the inverter circuit 210 is generating the high-frequency square-wave voltage VSQ. When the screw-in compact fluorescent lamp 120 is first powered up, the inverter power supply 240 is operable to generate the inverter supply voltage VINV before the control power supply 250 begins to produce the control supply voltage Vcc- After the inverter power supply 240 is generating the inverter supply voltage VINV, the inverter circuit 210 is able to begin generating the high-frequency square-wave voltage VSQ, such that the control power supply 250 is able to draw current from the resonant tank circuit 220. Accordingly, the control power supply 250 then produces the control supply voltage Vcc to power the control circuit 230.

[0042] The screw-in compact fluorescent lamp 120 further comprises an over-voltage protection (OVP) circuit 260, which provides an OVP control signal VOVP to the inverter circuit 210 for protecting the lamp tube 122 and the load regulation circuit 130 during over-voltage conditions. A lamp current sense circuit 270 is coupled in series with the lamp tube 122 to conduct the lamp current ILAMP and to generate a lamp current control signal VILAMP

representative of a magnitude of the lamp current ILAMP- The screw-in compact fluorescent lamp 120 further comprises a rectifier circuit 280 (e.g., a full-wave rectifier diode bridge as shown in Fig. 5), which has AC terminals coupled to receive the phase-control voltage Vpc from the filter network 200 and DC terminals for providing a rectified voltage VRECT-

[0043] The rectified voltage VRECT is received by an artificial load circuit 290 for conducting an artificial load current IART through the dimmer switch 30 in addition to the load current ILOAD conducted by the bus capacitors CBI, CB ₂ when the bus capacitors are charging. If the dimmer switch 30 includes a triac for generating the phase-control voltage Vpc, the artificial load circuit 290 is able to conduct enough current to ensure that the magnitude of the total current conducted through the triac of the dimmer switch exceeds the rated latching and holding currents of the triac. In addition, the artificial load circuit 290 may conduct a timing current if the dimmer switch 30 comprises a timing circuit and may conduct a charging current if the dimmer switch comprises a power supply, such that these currents need not be conducted through the load regulation circuit 130 and do not affect the intensity of the lamp tube 122. [0044] The artificial load circuit 290 comprises a current-passing switch, e.g., a FET

Q282, coupled in series with a resistor R284 (e.g., having a resistance of approximately 12.4 Ω), where the series combination of the drain-source junction of the FET Q282 and the resistor R284 is coupled across the DC terminals of the rectifier circuit 280. The gate of the FET Q282 is coupled to the drain of the FET Q282 through a resistor R285 (e.g., having a resistance of approximately 1 ΜΩ), such that the FET Q282 is rendered conductive and conducts the artificial load current IART when the magnitude of the rectified voltage VRECT increases from

approximately zero volts to exceed a turn-on threshold (e.g., approximately 4 volts).

Accordingly, the FET Q282 is rendered conductive, such that the artificial load circuit 290 conducts the artificial load current IART through the dimmer switch 30 after the triac is rendered conductive (if the dimmer switch is using the forward phase-control dimming technique), or shortly after the beginning of each half-cycle (if the dimmer switch is using the reverse phase- control dimming technique). The artificial load circuit 290 further comprises an NPN bipolar junction transistor Q286 having a base-emitter junction coupled across the resistor R284 and a collector coupled to the gate of the FET Q282. The transistor Q286 controls the FET Q282 in the linear region to provide over-current protection of the FET Q282 when the voltage across the resistor R284 exceeds the rated base-emitter voltage of the transistor Q286 (e.g., approximately 0.7 volt).

[0045] The artificial load circuit 290 also comprises an NPN bipolar junction transistor

Q288 having a collector coupled to the gate of the FET Q282, and a zener diode Z292, having, for example, break-over voltage VBR of approximately 13.2 volts. Three resistors R294, R295, R296 are coupled in series between the DC terminals of the rectifier circuit 180 and have, for example, resistances of approximately 22 kQ, 470 kQ, and 150 kQ, respectively. The zener diode Z292 is coupled between the base of the transistor Q288 and the junction of the resistors R295, R296. A capacitor C298 is coupled across the resistor R296 and has, for example, a capacitance of approximately 1000 pF. When the magnitude of the rectified voltage VRECT exceeds a turn-off threshold (e.g., approximately 60 volts), the zener diode Z292 conducts current into the base of the transistor Q288. Accordingly, the transistor Q288 is rendered conductive and the FET Q282 is rendered non-conductive, such that the artificial load circuit 290 stops conducting the artificial load current IART- If the dimmer switch 30 is using the forward phase-control dimming technique, the capacitor C298 provides some delay, such that the artificial load circuit 290 conducts the artificial load current IART for a period of time after the triac is rendered conductive.

[0046] The artificial load circuit 290 also provides a phase-control input control signal Vpc-iN to the control circuit 230. Specifically, the artificial load circuit 290 comprises a PNP bipolar junction transistor Q299 having a collector coupled to the control circuit 230 for providing the phase-control input control signal VPC-IN- The emitter-base junction of the transistor Q299 is coupled across the resistor R294, such that the transistor Q299 is rendered conductive shortly after the triac is rendered conductive (with forward phase-control dimming) or shortly after the beginning of each half-cycle (with reverse phase-control dimming). The control circuit 230 uses the phase-control input control signal VPC-IN to determine the duty-cycle DCpc of the phase-control voltage Vpc (i.e., the conduction time TCON of the bidirectional semiconductor switch 32 of the dimmer switch 30). The control circuit 230 determines the present magnitude of the lamp current ILAMP from the lamp current control signal VILAMP received from the lamp current sense circuit 270. The control circuit 230 then adjusts the operating frequency fop of the square wave voltage VSQ to control the magnitude of the lamp current ILAMP to a desired level that is dependent upon the duty-cycle DCpc of the phase-control voltage Vpc to thus control the intensity of the lamp tube 122 to the target intensity LTRGT-

[0047] Fig. 6 is a simplified schematic diagram of the electrical circuitry of the screw-in compact fluorescent lamp 120 showing the inverter circuit 210, the resonant tank circuit 220, the inverter power supply 240, the control power supply 250, and lamp current sense circuit 270 in greater detail. The inverter circuit 210 comprises first and second series-connected switching devices (e.g., FETs Q212, Q214) and an inverter control circuit 216, which may comprise, for example, an inverter control integrated circuit (IC) U300 (Fig. 5), such as part number FAN7532, manufactured by Fairchild Semiconductor Incorporated. The inverter control IC U300 of the inverter control circuit 216 is powered by the inverter supply voltage VLNV and controls the FETs Q212, Q214 in response to the drive control signal VDR received from the control circuit 230 as will be described in greater detail below with reference to Fig. 7. The inverter control circuit 216 renders the FETs Q212, Q214 conductive and non-conductive on a complementary basis (such that only one of the FETs is conducting at a time) using a constant duty cycle (e.g.,

approximately 50%). When the first FET Q212 is conductive, the output of the inverter circuit 210 is pulled up towards the bus voltage VBUS- When the second FET Q214 is conductive, the output of the inverter circuit 210 is pulled down towards circuit common. The magnitude of the lamp current ILAMP conducted through the lamp tube 122 is controlled by adjusting the operating frequency fop of the high-frequency square wave voltage VSQ generated by the inverter circuit 210.

[0048] The resonant tank circuit 220 comprises a resonant inductor L222 adapted to be coupled in series between the inverter circuit 210 and the lamp tube 122, and a resonant capacitor C224 adapted to be coupled in parallel with the lamp tube. For example, the inductor L222 may have an inductance of approximately 1.4 mH, while the resonant capacitor C224 may have a capacitance of approximately 1.2 nF, such that resonant tank circuit is characterized by a resonant frequency of approximately 110 kHz. The resonant tank circuit 220 further comprises a DC-blocking capacitor C226 that has a capacitance of, for example, approximately 2.7 nF and operates to minimize the DC component of the lamp current ILAMP conducted through the lamp tube 122.

[0049] When the lamp tube 122 is not illuminated and the control circuit 230 receives a command to turn the lamp tube on (from the phase-control voltage Vpc), the control circuit first preheats filaments 228 A, 228B of the lamp tube 122 and then attempts to strike the lamp tube. The load regulation circuit 130 may comprise, for example, two filament windings 229A, 229B that are magnetically coupled to the resonant inductor L222 and electrically coupled to the respective filaments 228A, 228B for generating filament voltages for heating the filaments of the lamp tube 122 prior to striking the lamp. To preheat the filaments 228A, 228B, the inverter circuit 210 controls the operating frequency fop of the square wave voltage VSQ to a preheat frequency fpu (e.g., approximately 150 kHz) for a preheat time period T _PH (e.g., approximately 700 msec). An example of a ballast having a circuit for heating the filaments of a fluorescent lamp is described in greater detail in U.S. Patent No. 7,586,268, issued September 8, 2009, titled APPARATUS AND METHOD FOR CONTROLLING THE FILAMENT VOLTAGE IN AN ELECTRONIC DIMMING BALLAST, the entire disclosure of which is hereby incorporated by reference.

[0050] The inverter power supply 240 receives the half-bus voltage VHB developed across the first bus capacitor CBI and generates the inverter supply voltage VL V across a storage capacitor C242 (e.g., having a capacitance of approximately 1 μΡ). The inverter power supply 240 comprises a simple zener-regulated power supply having a zener diode Z243, which is coupled across the storage capacitor C242 and may have, for example, a break-over voltage of approximately 13.2 volts. When the screw-in compact fluorescent lamp 120 is first powered up, the storage capacitor C242 is able to charge by conducting a small trickle current from the AC power source 15 through a resistor R244 (e.g., having a resistance of approximately 392 kQ) until the inverter supply voltage VINV is developed across the storage capacitor C242. After the magnitude of the inverter supply voltage VINV exceeds the rated operating voltage of the inverter control IC U300 of the inverter control circuit 216, the inverter control IC begins to control the FETs Q212, Q214 to be conductive and non-conductive to generate the square wave voltage VSQ.

[0051] The inverter power supply 240 further comprises a snubber capacitor C245 that has, for example, a capacitance of approximately 470 pF and provides a path for charging the storage capacitor C242 after the inverter control circuit 216 begins generating the square wave voltage VSQ. The snubber capacitor C245 is coupled between junction of the two FETs Q212, Q214 and the storage capacitor C242 through a diode D246 and a resistor R246 (e.g., having a resistance of approximately 5.6 Ω). The storage capacitor C242 is able to charge when the first FET Q212 is conductive (i.e., when the square-wave voltage VSQ is being pulled high towards the bus voltage VBUS). When the second FET Q214 is conductive and the square-wave voltage VSQ is being pulled low towards circuit common, the snubber capacitor C245 is able to discharge through the second FET Q214 and a diode D248. Accordingly, after the inverter control circuit 216 begins generating the square wave voltage VSQ, the inverter power supply 240 is operable to generate the inverter supply voltage VINV by conducting current through the snubber capacitor C245 rather than conducting current through the resistor R244, which would needlessly dissipate an excessive amount of power.

[0052] The control power supply 250 comprises a linear regulator, for example, an adjustable linear regulator U252, such as part number LM317L, manufactured by Fairchild Semiconductor Incorporated. The control power supply 250 comprises a winding 254 magnetically coupled to the resonant inductor L222 of the resonant tank circuit 220 for generating an electromagnetically-coupled voltage, such that the linear regulator U252 is able to draw current from the resonant tank circuit through a diode D255 when the inverter control circuit 216 is generating the square- wave voltage VSQ. A capacitor C256 is coupled across the input of the linear regulator U252 and has, for example, a capacitance of approximately 0.1 μΡ. A first resistor R258 is coupled between the output of the adjustable linear regulator U252 and the adjustment pin of the linear regulator, while a second resistor R259 is coupled between the adjustment pin and circuit common. For example, the first and second resistors R258, R259 may have resistances of approximately 475 Ω and 5.23 kQ, respectively, such that the control supply voltage Vcc at the output of the linear regulator U252 has a nominal magnitude of approximately 15 volts.

[0053] Fig. 7 is a simplified schematic diagram of the inverter circuit 210 and the OVP circuit 260. As previously mentioned, the inverter control circuit 216 includes the inverter control IC U300, which is powered by the inverter supply voltage VINV- The inverter control IC U300 is directly coupled to the gates of the FETs Q212, Q214 for controlling the FETs to be conductive and non-conductive (via pins 16 and 13). The inverter control circuit 216 comprises a preheat-frequency-set resistor R _PH , which is coupled to a preheat-frequency-set input (pin 7) of the inverter control IC U300 for setting the preheat frequency fpH- The preheat-frequency-set resistor R _PH may have a resistance of, for example, approximately 27 kQ, such that the preheat frequency fpn is approximately 150 kHz. The inverter control circuit 216 also comprises a preheat-time-set capacitor CPH, which is coupled to a preheat-time-set input (pin 5) of the inverter control IC U300 for setting the length of the preheat time period T _PH . For example, the preheat-time-set capacitor CPH may have a capacitance of approximately 0.33 μΡ, such that the preheat time period TPH is approximately 700 msec.

[0054] During the preheat time period T _PH , a voltage VRPH generated across the preheat-frequency-set resistor R _PH (i.e., at pin 7) is maintained constant, while a voltage VCPH generated across the preheat-time-set capacitor CPH (i.e., at pin 5) increases in magnitude with respect to time from approximately zero volts. When the voltage VCPH across the preheat-time- set capacitor CPH exceeds a preheat voltage threshold VPH at the end of the preheat time period TPH, the inverter control IC U300 then controls the operating frequency fop to attempt to strike the lamp tube 122. The voltage VRPH across the preheat-frequency-set resistor R _PH and the voltage VCPH across the preheat-time-set capacitor CPH are also provided to the control circuit 230, such that the control circuit is operable to properly control the inverter control IC U300 during the preheat time period T _PH as will be described in greater detail below.

[0055] The inverter control circuit 216 comprises an operating-frequency-set resistor RT coupled to an frequency-set-resistor input (pin 8) of the inverter control IC U300 and an operating-frequency-set capacitor CT coupled to a frequency-set-capacitor input (pin 6) for setting the operating frequency fop of the square-wave voltage VSQ when the lamp tube 122 is illuminated (i.e., after the lamp tube has been struck). For example, the operating-frequency-set resistor R _T may have a resistance of approximately 30 kQ and the operating-frequency-set capacitor CT may have a capacitance of approximately 330 pF, such that a default operating frequency of the square- wave voltage VSQ is approximately 1 10 kHz.

[0056] The inverter control circuit 216 further comprises an NPN bipolar junction transistor Q310 having a collector-emitter junction coupled between the frequency-set-resistor input of the inverter control IC U300 and circuit common through a resistor R312 (e.g., having a resistance of approximately 10 kQ). The base of the transistor Q310 is coupled to receive the drive control signal VDR from the control circuit 230. The drive control signal VDR has a DC magnitude that is representative of a target operating frequency fm _G T to which the operating frequency fop should be controlled to control the intensity of the lamp tube 122 to the target intensity LTRGT- The transistor Q310 is controlled to operate in the linear region, such that the transistor Q310 provides a controllable impedance between the frequency-set-resistor input of the inverter control IC U300 and circuit common in response to the DC magnitude of the drive control signal VDR. Accordingly, the control circuit 230 is operable to adjust the operating frequency fop of the square-wave voltage VSQ by controlling the impedance provided by the transistor Q310 and thus the current conducted through the frequency-set-resistor input of the inverter control IC U300.

[0057] The inverter control circuit 216 receives the OVP control signal VOVP from the

OVP circuit 260. Specifically, the OVP control signal VOVP is coupled to an open lamp protection (OLP) input (pin 10) of the inverter control IC U300 through a resistor R320 (e.g., having a resistance of approximately 10 kQ), and is coupled to circuit common through a capacitor C322 (e.g., having a capacitance of approximately 0. 1 μΡ). The OVP circuit 260 comprises a voltage divider having resistors R261 , R262 for scaling the magnitude of the lamp voltage VLAMP down to a magnitude that is appropriate to be received by the inverter control IC U300. For example, the resistors R261, R262 may have resistances of approximately 1 ΜΩ and 25.5 kQ, respectively. The junction of the resistors R261 , R262 is coupled to a capacitor C264 (e.g., having a capacitance of approximately 0. 1 μΡ) through a diode D265. The junction of the capacitor C264 and the diode D265 is coupled to the inverter control circuit 216 through a zener diode Z266 for generating the OVP control signal VOVP, which is coupled to circuit common through a resistor R268 (e.g., having a resistance of approximately 100 kQ). For example, the zener diode Z266 may have a break-over voltage of approximately 13.2 volts. [0058] During normal operation, the voltage at the OLP input of the inverter control IC

U300 remains low (i.e., at approximately circuit common). However, in the event of an overvoltage condition across the lamp tube 122, the zener diode Z266 begins to conduct, such that the voltage at the OLP input of the inverter control IC U300 increases in magnitude. When the magnitude of the voltage at the OLP input exceeds an OLP threshold voltage of the inverter control IC U300 (e.g., approximately 2 volts), the inverter control IC disables the outputs (i.e., pins 13 and 16) such that the FETs Q212, Q214 are rendered non-conductive and the lamp tube 122 is not illuminated until the control circuit 230 controls the inverter circuit 210 to attempt to restrike the lamp tube once again.

[0059] Fig. 8 is a simplified schematic diagram of the control circuit 230. The control circuit 230 comprises a two-speed phase-to-DC converter circuit 400 for converting the phase-control input control signal VPC-IN to a DC target voltage VTRGT that is representative of the duty-cycle DCpc of the phase-control voltage Vpc and thus the target intensity L _T RGT of the lamp tube 122. The DC target voltage VTRGT is amplified by a non-inverting amplifier circuit 420 to generate an amplified target voltage VA-TRGT, such that the magnitude of the amplified target voltage VA-TRGT is within a correct range to be received by an error amplifier circuit 430. For example, the DC target voltage VTRGT may be in the range of approximately 1-4 volts, while the amplified target voltage VA-TRGT is in the range of approximately 0.5-6.5 volts. A non-linear amplifier circuit 440 receives the lamp current control signal VILAMP from the lamp current sense circuit 270 and generates an amplified lamp current signal VA-ILAMP- The error amplifier circuit 430 receives the amplified target voltage VA-TRGT and the amplified lamp current signal VT-ILAMP and generates the drive control signal VDR to adjust the operating frequency fop of the inverter circuit 210, so as to minimize the error between the amplified target voltage VA-TRGT and the amplified lamp current signal VT-ILAMP-

[0060] The phase-to-DC converter circuit 400 comprises a voltage divider having two resistors R401, R402 for scaling down the phase-control input control signal VPC-IN- For example, the resistors R401, R402 may have resistances of approximately 1 ΜΩ and 47 kQ, respectively. Next, the phase-to-DC converter circuit 400 generates a switched voltage Vs that has a duty-cycle approximately equal to the duty cycle DCpc of the phase-control input control signal VPC-IN- Specifically, the junction of the resistors R401, R402 is coupled to the base of a first NPN bipolar junction transistor Q404 that has a collector-emitter junction coupled between the control supply voltage Vcc and circuit common through a resistor R405 (e.g., having a resistance of approximately 220 kQ). The junction of the collector of the transistor Q404 and the resistor R405 is coupled to the base of a second NPN bipolar junction transistor Q406 that has a collector-emitter junction coupled between the control supply voltage Vcc and circuit common through two resistors R408, R409 (e.g., having resistances of approximately 40 kQ and 1 kQ, respectively). Accordingly, the switched voltage Vs is generated at the collector of the second transistor Q406.

[0061] When the magnitude of the phase-control input control signal VPC-IN is approximately zero volts (i.e., when the bidirectional semiconductor switch 32 of the dimmer switch 30 is non-conductive), the first transistor Q404 is rendered non-conductive, such that the second transistor Q406 is rendered conductive and the switched voltage Vs is pulled low towards circuit common. When the magnitude of the phase-control input control signal VPC-IN is greater than a phase-control threshold, e.g., approximately 15 volts (i.e., when the bidirectional semiconductor switch 32 of the dimmer switch 30 is conductive), the first transistor Q404 is rendered conductive, such that the second transistor Q406 is rendered non-conductive and the switched voltage Vs is pulled high towards control supply voltage Vcc through the resistors R408, R409.

[0062] A ramp voltage VR is generated across a capacitor C410 (e.g., having a capacitance of approximately 0.22 μΡ) in response to the square-wave voltage Vs. When the second transistor Q406 is non-conductive, the capacitor C410 is able to charge towards the control supply voltage Vcc through the resistor R408, such that the magnitude of the ramp voltage VR increases with respect to time while the switched voltage Vs is high. When the second transistor Q406 is conductive, the capacitor C410 is able to discharge through the resistor R409, such that the magnitude of the ramp voltage VR decreases at a second rate that is much faster than the first rate at which the ramp voltage increases in magnitude. Accordingly, the ramp voltage VR is generated across the capacitor C410 and has a duty cycle equal to approximately the duty cycle of the phase-control input control signal VPC-IN- Next, the ramp voltage VR is filtered by a first filter circuit, e.g., a resistor-capacitor (RC) circuit (including a resistor R411 and a capacitor C412), to generate a filtered voltage VF. For example, the resistor R411 has a resistance of approximately 220 kQ and the capacitor C412 has a capacitance of approximately 0.22 μΡ, such that the first RC circuit has a time constant XRO of approximately 48.4 msec. [0063] The filtered voltage VF from the first RC circuit is then filtered by a second RC circuit (having a resistor R413 and a capacitor C414) to generate the target voltage VTRGT- TWO diodes D416, D418 are coupled in anti-parallel connection across the resistor R413. For example, the resistor R413 has a resistance of approximately 2.2 ΜΩ and the capacitor C414 has a capacitance of approximately 0.22 μΡ, such that a time constant XRC2 of the second RC circuit has a nominal value of approximately 484 msec (i.e., approximately 10 times slower than the first RC circuit). The magnitude of the target voltage VTRGT is a function of the square of the conduction time T _C ON of the bidirectional semiconductor switch 32 of the dimmer switch 30, i.e., VTRGT =/ (TCON ² )- Accordingly, the control circuit 230 is operable to adjust the intensity of the lamp tube 122 in response to the duty-cycle DCpc of the phase-control voltage Vpc according to a "square-law" dimming curve. As a result, the control circuit 230 provides finer tuning of the intensity of the lamp tube 122 near the low-end intensity L _L E, such that larger variations in the conduction time T _C ON of the bidirectional semiconductor switch 32 result in smaller changes in the intensity of the lamp tube 122 near the low-end intensity L _L E-

[0064] Voltage fluctuations in the AC mains line voltage VAC of the AC power source 15 or noise on the AC mains line voltage VAC can cause the duty-cycle DCpc of the phase-control signal Vpc and the magnitude of the filtered voltage VF to vary slightly. Therefore, the slow nominal value XRC2-NOM of the time constant XRC2 of the second RC circuit provides enough filtering such that the target voltage VTRGT is not responsive to changes in the filtered voltage VF that are less than a predetermined threshold, e.g., approximately the forward voltage of the diodes D416, D418, i.e., a diode drop (e.g., approximately 0.7 volts).

[0065] However, changes in the target intensity LTRGT at the dimmer switch 30 that result in the dimmer switch changing the duty cycle DCpc of the phase-control signal Vpc cause the magnitude of the filtered voltage VF to change by greater amounts and at faster rates than the voltage fluctuations and noise of the AC mains line voltage VAC- Therefore, when the magnitude of the filtered voltage VF changes by more than the forward voltage of the diodes D416, D418, one of the diodes D416, D418 will begin to conduct such that the magnitude of the target voltage VTRGT across the capacitor C414 changes quickly in response to changes in the target intensity LTRGT- Specifically, the first diode D416 is operable to conduct current into the capacitor C414 when the target intensity L _T RGT increases, such that the time constant XRC2 has a fast value XRC2- FAST that is less than the nominal value XRC2-NOM- In addition, the capacitor C414 is operable to discharge through the diode D418 with the fast value XRC2-FAST when the target intensity L _T RGT decreases. Accordingly, the two-speed phase-to-DC converter circuit 400 is able to filter out changes in the duty cycle DCpc of the phase-control voltage Vpc due to voltage fluctuations and noise of the AC mains line voltage VAC while still providing a fast response as a result of changes in the target intensity LTRGT-

[0066] Each of the diodes D416, D418 stops conducting when the difference between the magnitudes of the target voltage VTRGT and the filtered voltage VF falls below approximately the forward voltage of the respective diode. After the diodes D416, D418 stop conducting, the magnitude of the target intensity L _T RGT will slowly change to be equal to the magnitude of the filtered voltage VF (according to the nominal time constant XRC2-NOM) until the magnitudes of the voltages are equal. This results in a slow fading of the intensity of the lamp tube 122 at the end of a change in the target intensity LTRG, which provides a pleasant, soft effect on a human eye that is observing the change in the intensity of the lamp tube.

[0067] The non-inverting amplifier circuit 420 comprises an operational amplifier ("op amp") U421, such as, for example, part number LM2902PWR, manufactured by National Semiconductor Corporation. The target voltage VTRGT is coupled to the non-inverting input of the op amp U421, while an offset voltage VOFF is coupled to the inverting input of the op amp through a resistor R422 (e.g., having a resistance of approximately 160 kQ). The offset voltage VOFF is generated by a voltage divider that is coupled between the control supply voltage Vcc and circuit common, and includes two resistors R424, R425. For example, the resistors R424, R425 may have resistances of approximately 33 kQ and 5 kQ, respectively, such that the offset voltage VOFF has a magnitude of approximately 2 volts. The output of the op amp U421 is coupled the inverting input via the parallel combination of a resistor R426 (e.g., having a resistance of approximately 150 kQ) and a capacitor C428 (e.g., having a capacitance of approximately 0.22 μΡ). The magnitude of the amplified target voltage VA-TRGT ranges from approximately 0.5 to 6.5 volts as the magnitude of the target voltage VTRGT ranges from approximately 1 to 4 volts.

[0068] The non-linear amplifier circuit 440 receives the lamp current control signal

VILAMP from the lamp current sense circuit 270, which is shown in Fig. 6. The lamp current control signal VILAMP is generated across the parallel combination of a resistor R272 (e.g., having a resistance of approximately 4Ω) and a capacitor C274 (e.g., having a capacitance of approximately 2.2 μΡ). During the negative half-cycles of the lamp current ILAMP, the lamp current is conducted through a diode D276 and the parallel combination of the resistor R272 and the capacitor C274. During the positive half-cycles of the lamp current ILAMP, the lamp current is conducted through a diode D278 and is not conducted through the parallel combination of the resistor R272 and the capacitor C274. Accordingly, the lamp current control signal VILAMP has a negative magnitude that is representative of the magnitude of the lamp current ILAMP during the negative half-cycles of the lamp current. Since the resistor R272 only conducts the lamp current ILAMP every other half-cycle, the resistor R272 dissipates half of the amount of power that would be dissipated if the resistor R272 conducted the lamp current each half-cycle.

[0069] Referring back to Fig. 8, the non-linear amplifier circuit 440 comprises an op amp U441 having a non- inverting input coupled to circuit common and an inverting input coupled to receive the lamp current control signal VILAMP through a resistor R442 (e.g., having a resistance of approximately 1 kQ). The output of the op amp U441 is coupled to the inverting input through a resistor R444 (e.g., having a resistance of approximately 68.1 kQ). The nonlinear amplifier circuit 440 further comprises a PNP bipolar junction transistor Q445 and a resistor R446 (e.g., having a resistance of approximately 27 kQ). The series combination of the collector-emitter junction of the transistor Q445 and the resistor R446 is also coupled between the inverting input and the output of the op amp U441. A capacitor C448 is coupled is parallel with the resistor R446 and may have, for example, a capacitance of approximately 470 pF.

[0070] When the magnitude of the lamp current ILAMP is less than a current threshold

(e.g., approximately 100 mA), the magnitude of the amplified lamp current signal VA-ILAMP is less than approximately the rated emitter-base voltage of the transistor Q445. At this time, only the resistor R444 is coupled between the inverting input and the output of the op amp U441, such that the non-linear amplifier circuit 440 is characterized by a first gain ! (e.g., approximately - 68). However, when the magnitude of the lamp current ILAMP is greater than the current threshold, the transistor Q445 is rendered conductive, such that the resistor R446 is coupled in parallel with the resistor R444 between the inverting input and the output of the op amp U441. Accordingly, above the current threshold, the non-linear amplifier circuit 440 is characterized by a second gain a ₂ that has a smaller magnitude than the first gain ! (e.g., approximately -25).

[0071] The error amplifier circuit 430 comprises an op amp U431 having a non-inverting input coupled to receive the amplified lamp current signal VA-ILAMP and an inverting input coupled to receive the amplified target voltage VA-TRGT through a resistor R432 (e.g., having a resistance of approximately 30 kQ). The error amplifier circuit 430 further comprises two capacitors C434, C435 (e.g., each having a capacitance of approximately 4.7 nF) and a resistor R436 (e.g., having a resistance of approximately 47 kQ). The capacitor C434 is coupled between the inverting input and the output of the op amp U431, while the series combination of the capacitor C435 and the resistor R436 is also coupled between the inverting input and the output of the op amp U431. The output of the op amp U431 is coupled to circuit common through a resister divider having two resistors R438, R439 (e.g., having resistances of 18.7 kQ and 6.8 kQ, respectively), where the drive control signal VDR is produced at the junction of the resistors R438, R439. The error amplifier circuit 430 operates to adjust the operating frequency fop of the inverter circuit 210 so as to minimize the error (i.e., the difference) between the amplified lamp current signal VA-ILAMP and the amplified target voltage VA-TRGT- For example, the error amplifier circuit 430 may be characterized by a cutoff frequency of greater than or equal to approximately 10 kHz (i.e., the control circuit 230 has a bandwidth greater than or equal to approximately 10 kHz).

[0072] As previously mentioned, the lamp tube 122 of the screw-in compact fluorescent lamp 120 of the embodiment of the present invention may be filled with the fill-gas mixture having a fill-gas pressure of approximately 2 Torr and a fill-gas ratio of approximately 85 : 15 argon to neon. Fig. 9 is an example V-I curve 500 (i.e., the plot of the magnitude of the lamp voltage VLAMP across the lamp tube 122 with respect to the magnitude of the lamp current ILAMP conducted through the lamp tube) according to the embodiment of the present invention. Fig. 9 also shows an example V-I curve 550 of the prior art screw-in compact fluorescent lamp 20 that has a fill gas of 100% argon at a pressure of approximately 4 Torr.

[0073] By combining the lamp tube 122 having the fill-gas mixture having a fill-gas ratio of approximately 85: 15 argon to neon at a pressure of approximately 2 Torr and the error amplifier 430 having a cutoff frequency of approximately 10 kHz, the V-I curve 500 of the screw-in compact fluorescent lamp 120 of the embodiment of the present invention is much "flatter" than the V-I curve 550 of the prior art screw-in compact fluorescent lamp 20 as shown in Fig. 9. In other words, the magnitude of the lamp voltage VLAMP does not changes as much with respect to the magnitude of the lamp current ILAMP as in the prior art screw-in compact fluorescent lamp 20. Therefore, the size of the components of the resonant tank circuit 220 may be smaller and the resonant tank circuit may operate more efficiently during normal operation. Since the filament windings 229 A, 229B are magnetically coupled to the resonant inductor L222, the flatter V-I curve 500 of the screw-in compact fluorescent lamp 120 of the embodiment of the present invention provides for more favorable magnitudes of the filament voltages across the dimming range of the lamp tube 122, particularly, near the near the low-end intensity L _L E when the magnitude of the lamp voltage VLAMP begins to decrease as the magnitude of the lamp current ILAMP decreases.

[0074] Figs. 10A and 10B show example waveforms of the DC bus voltage VBUS and the lamp current ILAMP, respectively, of the screw-in compact fluorescent lamp 120 of the embodiment of the present invention. Since the cutoff frequency of the error amplifier circuit 430 (i.e., approximately 10 kHz) is much greater than the frequency of the voltage ripple of the bus voltage VBUS (i.e., approximately 120 Hz), the control circuit 230 is able to adjust the operating frequency fop of the square-wave voltage VSQ (and thus the lamp current ILAMP) within a single half-cycle of the AC power source 15. Therefore, the operating frequency fop changes in magnitude in response to the voltage ripple of the bus voltage VBUS during each half-cycle and the lamp current ILAMP has an envelope IENV that is relatively constant or flat (as shown in Fig. 10B). It has been observed that this high speed operation of the error amplifier circuit 430 reduces flickering in the lamp tube 122, particularly when the lamp tube is cold and/or has just been started up. In addition, since the operating frequency fop of the inverter circuit 210 changes during each half-cycle even when the screw- in compact fluorescent lamp 120 is maintaining the target intensity L _T RGT of the lamp tube 122 constant, the compact fluorescent lamp does not generate EMI noise at specific frequencies, but spreads the frequency of the EMI noise out over a range, such that the peak magnitude of the noise is decreased.

[0075] Referring back to Fig. 8, the control circuit 230 further comprises a preheat adjustment circuit 450 that modifies the operation of the phase-to-DC converter circuit 400 while the load regulation circuit 130 is preheating the filaments 228 A, 228B of the lamp tube 122 (i.e., during the preheat time period TPH). The preheat adjustment circuit 450 comprises an op amp U452 having a non-inverting input coupled to receive the voltage VRPH across the

preheat-frequency-set resistor R _PH of the inverter control circuit 216, and an inverting input coupled to receive the voltage VCPH across the preheat-time-set capacitor CPH. The output of the op amp U452 is coupled to the capacitor C410 of the phase-to-DC converter circuit 400 through a resistor R454 (e.g., having a resistance of approximately 464 kQ). As previously mentioned, while the inverter control IC U300 is preheating the filaments 228A, 228B of the lamp tube 122, the voltage VRPH across the preheat-frequency-set resistor R _PH is maintained constant, while the voltage VCPH across the preheat-time-set capacitor CPH increases in magnitude with respect to time from approximately zero volts. During the preheat time period TPH, the op amp U452 injects current into the capacitor C410 of the phase-to-DC converter circuit 400, such that the magnitude of the target voltage VTRGT increases. Therefore, if the lamp tube 122 is being turned on to an intensity near the low-end intensity L _L E, the magnitude of the lamp current ILAMP will be large enough at the end of the preheat time period TPH to ensure that the arc current is established in the lamp tube when the lamp tube is struck. After the arc current has been properly established in the lamp tube 122, the op amp U452 stops injecting current into the capacitor C410 of the phase-to-DC converter circuit 400 and the intensity of the lamp tube is controlled in response to the target intensity L _T RGT determined from the duty-cycle DCpc of the phase-control voltage Vpc.

[0076] Although the present invention has been described in relation to particular embodiments thereof, many other variations and modifications and other uses will become apparent to those skilled in the art. It is preferred, therefore, that the present invention be limited not by the specific disclosure herein, but only by the appended claims.