Patent Application
20090295300
The present disclosure relates generally to electronic lighting ballasts and, more particularly, to methods and apparatus for high efficiency ballasts for use with light emitting diode (“LED”) based light sources that can be effectively dimmed and configured to operate with a high power factor.
In the field of lighting, LEDs are emerging as a promising technology for generating light at high efficiency. Traditionally, LEDs have been used in consumer electronics as indicators (such as function indicators, power indicators, etc.). The development of LEDs that generate white light (as opposed to LEDs that produced red, green, or other light colors) allows LEDs to be used as potential general purpose lighting sources. While LEDs provide a relatively high lumens/watt, they are presently limited in the amount of power that can be converted into light. Unlike incandescent bulbs which convert very little of the input energy into light (about 90% of the energy input into an incandescent light bulb is used to generate heat), LEDs convert a high percentage of input power into light. Further, unlike fluorescent lamps and other forms of gas-discharge lamps, LEDs are solid state devices and do not rely on a glass or quartz bulb to contain gases (which often contain hazardous materials such as mercury) that are ionized. Finally, LEDs are individually smaller and more reliable than bulbs.
Traditionally, LEDs were limited in the power they could dissipate and many LEDs are still designed for relatively low power (conventional LEDs draw only 20 milli-amps and are rated at only 1/10 watt). Indeed, the prior development and incorporation of LEDs in many battery operated devices was based on their low power consumption, and hence their low power levels were not considered a limiting aspect, but a desirable aspect. However, recent advances to adapt LEDs as light sources have resulted in development of relatively high powered LEDs. A high power LED may be considered an LED capable of handling at least ½ watt, but LEDs are presently available that consume 6 or more watts of power. In comparison, a typical incandescent bulb is rated at 60-100 watts (with higher wattages readily available), and a compact fluorescent bulb is typically rated between 11 and 40 watts. These ranges are not absolute values, but represent typical ranges. Thus, while an LED maybe more efficient than an incandescent or fluorescent bulb in generating light, the total light output of a single LED is typically less than conventional light sources. In summary, while conventional light sources can handle greater amounts of power than individual LED light sources, they are less efficient.
Two approaches for providing more light using LEDs are possible. First, LEDs are available (and likely will be developed) to handle greater power, therefore each can individually generate more light than conventional LEDs. Second, a plurality of conventional LEDs can be used to function as a single light source. In the latter case, LED lighting panels or strips are commercially available that can comprise hundreds of LEDs functioning as a single light source.
LEDs are a form of diode and operate on a DC current. Typically, the voltage across an individual LED is relatively low, typically only several volts. It is well known that a simple circuit for limiting direct current in an LED can comprise a current limiting resistor connected to a DC voltage source that passes current through an LED. These circuits are relatively simple, but have the disadvantage that the resistor is a passive element and any energy dissipated through it is energy that is not converted into useful light. Hence, such systems are not energy efficient.
If LEDs are to become viable substitutes for conventional light sources (incandescent or gas-discharge bulbs), it would be desirable to be able to dim the LEDs. Various lighting applications require, or benefit from, dimming light sources. For example, to become a viable replacement for incandescent bulbs in certain residential applications, market requirements would dictate that LEDs be dimmable. In other applications, including so-called “daylight harvesting” applications, energy savings is achieved by dimming lights based on ambient lighting conditions. Thus, if natural daylight is sufficient in the desired area, the lighting source may be automatically dimmed. If natural daylight is insufficient, then the lighting levels are increased. This application is common in security lighting and energy savings applications.
Consequently, circuitry for controlling LED light sources in lighting applications requires an energy efficient circuit for providing current to one or more LEDs, but at the same time should provide dimming capability and efficient operation.
In addition, because conventional lighting frequently operates on household AC voltage, the control circuitry for LEDs should be able to operate using household power (e.g., 120 volts and 60 Hertz in the U.S., 240 volts and 50 Hertz in many other countries). This requires circuitry for converting AC to a lower level DC voltage. Again, this circuitry should be energy efficient, and should be compatible with dimming circuits.
However, a problem can arise when using conventional dimmers in certain type of lighting circuits. While many prior art dimmers operate fine with incandescent lamps having a minimum wattage, operating the same dimmers with ballasts can be problematic. Some dimmers state that 20 to 40 watts are required as a minimum load, and hence do not operate properly with lower rated loads. Because LEDs typically have a high efficiency and present a lower load (frequently less than 20 or 40 watts), LED light sources may not meet the minimum power required by a conventional dimmer. Other dimmers do not have this requirement, but they are more complex (and hence more costly). In other instances, ballasts for controlling a LED light source may require specially designed dimmers, which cannot be used with other lighting fixtures.
The ballast (e.g., the circuitry for controlling current through the LED) should also provide a favorable power factor (“pf”). The power factor has a range of between 0 and 1 and is generally defined as the relationship of the real power to the apparent power. In an electric power system, a load with low power factor draws more current than a load with a high power factor for the same amount of useful power transferred to the load. The higher current increases the energy lost in the distribution system, and requires at an aggregate level larger distribution wires and equipment by the distribution system. Because of the costs of larger equipment and wasted energy, electrical utilities will usually charge a higher rate to industrial or commercial customers having a low power factor. In summary, a low power factor in the lighting ballast causes inefficiency in the power distribution system and is undesirable.
An incandescent bulb typically has a very high power factor (better than pf=0.9), and is desirable in this respect. However, as noted, incandescent bulbs are not very efficient in converting incoming power into light. While gas-discharge lights such as fluorescent bulbs, are more efficient, the circuitry used to drive the bulb typically have a lower power factor (0.5-0.7). In this regard, they are undesirable. Thus, it would be desirable to have LEDs (which are very efficient) to have a high power factor. It is commonly accepted that for loads less than 100 watts, a high power factor is pf=0.9 or higher. For loads greater than 100 watts, a high power factor is p=0.95 or greater. Because LEDs are relatively low power, typically the former classification is used (e.g., a high power factor is pf=0.9 or higher).
Further, there is a practical benefit to having a ballast that can be easily and reliably manufactured using few parts than other ballasts, and which can be easily adapted for not only gas-discharge lamps, but also for use with LED light sources.
Therefore, there is a need for circuitry for controlling one or more LEDs that is energy efficient, allows dimming of the LEDs, and maintains a high power factor.
Methods and apparatus are disclosed for dimmable ballast circuits that operate with LED light sources. In one embodiment, a dimmable ballast circuit receives alternating voltage from a power source and provides rectified line voltage to a first node and a second node, wherein the power source provides a current alternating at a line frequency. The first node and the second node are connected to each other via a bypass capacitor that presents high impedance at the line frequency. The bypass capacitor filters high frequency noise and stores high frequency energy in order to provide current at a switching (high) frequency when discharged. Typically, the switching frequency is at least two orders of magnitude higher than the line frequency. This capacitor is small enough in capacitance value relative to the load and line operating frequency that it provides a relatively large reactance to the rectified AC input from the power source at the line frequency. A first switch is operable to selectively couple the first node where the rectified line voltage is provided to a resonant circuit. The resonant circuit has a resonant frequency and stores energy during a portion of the switching cycle thereby generating a voltage across a diode bridge to which a LED light source is connected. Once the threshold voltage of the LED light source is exceeded, current flows through the LED, and light is emitted. In one embodiment, a second switch is operable to selectively couple the resonant circuit to the second node while the first switch is opened. This allows energy stored in the resonant circuit to be substantially recycled within the resonant circuit to also generate light.
a and 1b illustrate one embodiment of a lighting ballast for a single LED.
a-2c illustrate voltage waveforms present in the lighting ballast of
a and 3b are current flow diagrams illustrating operation of the lighting ballast.
a and 6b illustrate voltage waveforms present in the tank circuit of one embodiment of a LED lighting ballast.
a-10b illustrates waveform associated with a ballast used with a phase control dimmer.
Methods and apparatus for dimmable ballasts for use with one or more LED are described herein. In the described examples, a dimmable ballast circuit, typically having a high power factor, is described that interfaces a power source with a light source comprising one or more LEDs. The disclosed dimmable ballasts include a high frequency filter capacitor to reduce high frequency energy from entering the power supply during its operation, allow operation of the ballast, and increase the efficiency of the ballast.
a and 1b together illustrate one embodiment of an electrical lighting ballast capable of operating on household power, which typically in the U.S. is 120 VAC/60 Hz. Other countries may operate using 240 VAC/50 Hz and suitable changes in the component values may be necessary and are within the knowledge of one of ordinary skill in the art. Although various embodiments herein are disclosed in terms of “household voltage,” or “household power,” these terms refer to any readily available line voltage at a line frequency, and does not preclude application to other commercial or industrial power sources. Thus, for example, the principles of the present invention could be adapted to other voltages and frequencies, such as the 400 Hz AC systems used in commercial aircraft. Hence, variations regarding the power source characteristics are possible, which may impact the precise values of various components used.
The embodiment of
Typically, the light source will be integrated in a non-user removable manner with the ballast and can be considered as part of the ballast. LEDs typically have a long life and are not expected to require replacement, but it is possible that in some embodiment, the LEDs (or the ballast) could be replaced separately from the light source. In other contexts herein, the ballast may be described as being the circuitry for providing current to the light source, and thus excludes the LED(s). However, whether the LED is considered part of the ballast as used herein will be clear from the context, or in many cases, is not material to the explanation of the operation of the present invention.
The first section discussed is the power input section 2 in
During operation, the power input section essentially receives and provides 120 VAC at a 60 Hz line frequency from a power source (usually obtain by a receptacle or otherwise wired to a power distribution point in a building) to the input of the power rectifier section 4. The filter components aid in reducing noise from being introduced into the power line from the remainder of the ballast, and provide safety mechanisms to limit potential damage from high current or voltage.
The input AC voltage from the power input section is provided to the rectifier section 4. The rectifier comprises a full wave bridge diode assembly comprising diodes 104a-104d rectifying the AC voltage to produce an unfiltered rectified DC voltage. These diodes can comprise 1 amp, 400 v 1N4001 diodes, although other embodiments can utilize a full wave bridge in the form of a single component. Unlike prior art ballasts which often incorporate a “smoothing” capacitor in the form of an electrolytic capacitor, the embodiment in
The voltage waveform of
The ballast circuit also includes a voltage regulator section 6. This is sometimes referred to as a housekeeping supply circuit since it provides power necessary to maintain operation of the IC driver chip 132. The voltage regulator is connected to node 50 and 55, and receives power from the output of the full wave bridge. Voltage regulator 6 generates a substantially constant voltage that exceeds a minimum threshold (e.g., 10 volts, etc.) to provide power to the integrated circuit driver 132. Because the voltage at nodes 50, 55 is not filtered by a smoothing capacitor, a regulator is required to provide a steady input voltage to the driver. Recall that the voltage waveform from the rectifier section 4 has at each half cycle a “valley” wherein the voltage drops to zero or near-zero, albeit for a short time. If the voltage to the IC were to fall to zero (or near zero) volts during this time, the driver chip may cease to function. In certain cases, there may be sufficient charge stored in the IC itself to overcome these brief valleys in the supply voltage. However, when using the ballast with a dimmer, the period which the input voltage is zero increases in duration, and the IC would be unable to continue functioning. Thus, a voltage regulator is incorporated.
In the illustrated embodiment, voltage regulator section 6 is implemented using an NMOS transistor 110 that is connected to the first node 50 via a resistor 108, which in one embodiment is 220 ohms. The drain of NMOS transistor 110 is connected to its respective gate via a resistor 106, which in one embodiment is 1M ohms. The gate of NMOS transistor 110 is further connected to a collector of a transistor 116 via an optional resistor 112, which in one embodiment is 1 k ohms, which has its respective base connected to the anode of a zener diode 114, which in one embodiment is a 14 v zener diode. Resistor 112 reduces the gain of the transistor thereby reducing possibility of oscillations in transistor 110. The cathode of zener diode 114 is connected to the source of NMOS transistor 110.
In addition, the base of transistor 116 is connected to second node 55 via resistor 120 which is one embodiment is a 10 k ohms, and its emitter is connected to the second node 55 via a resistor 118, which in one embodiment is 1 k ohms. In the example of
Referring to the IC driver 132, voltage regulator section 6 provides the substantially constant (i.e., regulated) voltage via diode 124, which also isolates voltage regulator 6 from driver 132. Stated differently, diode 124 prevents current from flowing from capacitor 129 into regulator 6 when the voltage of the first node 50 falls below the voltage stored in capacitor 129. In the embodiment of
An alternate embodiment of the voltage regulator section 6 is possible. One alternative embodiment that can be used if the ballast is not to be dimmed is shown in
This arrangement requires fewer parts for the voltage regulator embodiment of
The driver circuit 132 is configured to generate a signal that alternately actuates one of the transistors 144 and 148 at the switching frequency, which is much higher than the line frequency. In particular, during the first half (or a portion thereof) of a single cycle of the switching frequency, the high side output (HO) of the driver circuit 132 produces a high side pulse to turn on transistor 144 while transistor 148 is turned off. Typically, the high side pulse has a duration that does not exceed half of the time period of a cycle of the switching frequency. When the driver circuit 132 turns on transistor 144, the transistor 144 couples the node 50 to the resonant circuit 245 via a low impedance path.
Typically the switching frequency is 20 kHz or higher, and it is typically at least two orders of magnitude greater than the line frequency. Thus, reference herein to the “low frequency” refers to the line frequency, whereas reference to the “high frequency” refers to the switching frequency. In certain embodiments, the driver IC may be an International Rectifier® IR2153 self Oscillation Half-Bridge Driver integrated circuit. In other embodiments, a 555 timer IC or other pulse width generator circuit (including processor based) may be used to generate the signals for driving the switching transistors in switching section 10.
In the illustrated embodiment of
In the illustrated example, the resistance value of the resistor 126 and the capacitance value of the capacitor 124 configure the driver circuit 132 to produce pulses at a frequency in the range of approximately 20 to 100 KHz. Specifically, the pulses are alternately produced by driver circuit 132 and are output via the high side gate driver output (HO) and the low side gate driver output (LO). Stated differently, during the first half cycle of a period of the switching frequency (i.e., the half of the time period for a single switching cycle), the high side gate driver output of the driver circuit 132 produces a pulse. During the second half cycle of the period (i.e., the low side of the cycle) of the switching frequency, the low side gate driver output of the driver circuit produces a pulse. Typically, there is a dead time between pulses when neither transistor is turned on, e.g., the time after the first pulse ends and before the second pulse begins.
In the embodiment of
The switching section 10 comprises transistors 144 and 148 and are typically both implemented using vertical N-Channel metal oxide semiconductor (NMOS) field effect transistors, although one of ordinary skill in the art would know that these transistors can be implemented by any other suitable solid state switching device (e.g., a P-channel metal oxide field effect transistor, an insulated gate bipolar transistor (IGBT), a lateral N-channel mode MOS transistor, a bipolar transistors, a thyristor, gate turn off (GTO) device, etc.).
The IC driver 8 and switching section 10 form a half-bridge switching topology that is implemented to provide energy at output nodes 151 and 153, which in turn provide power to the resonant circuit portion 14 of “tank circuit” 150. It is desirable that the transistors switch at a zero-current or zero voltage condition so as to minimize the stress on the components, which impacts their longevity, and also the efficiency.
To form the half-bridge topology, the drain of the first transistor 144 is connected to the first node 50 and the source of the second transistor 148 is connected to the second node 55. Thus, the voltage present on node 50 and the drain of the first transistor 144 is the rectified voltage waveform 200 shown in
The nodes 151 and 153 represent the output of the switching section. Thus, the square waves 260 of
The final portion of the main ballast portion 150 is the bypass capacitor portion 12. This section comprises a single capacitor, termed the “bypass capacitor” herein, that is connected to nodes 50 and 55; specifically, across the outputs of the full wave bridge. Thus, the voltage present at the output of the full wave bridge section 2 is the same voltage across the terminals of the bypass capacitor. The bypass capacitor is a high frequency energy storage device, such as a polypropylene capacitor 102. It is typically not an electrolytic capacitor, since these are typically unsuitable for high-frequency operation. The bypass capacitor should not be confused with a “smoothing” electrolytic capacitor similarly positioned across the output of a full wave bridge rectifier in found the prior art, but which performs a different function. In the example of
The reactance is defined by the following formula in Equation 1:
In the case for a ballast operating at a switching frequency of 40 kHz, a 1 μF capacitor typically used and would present an reactance of about 4 ohms. However, this same capacitor would have a reactance at the line frequency of 60 Hz of about 2653 ohms. The line frequency (60 Hz) is generally fixed by the power source provider and thus a high impedance is presented by the bypass capacitor at the line frequency, typically greater than 1500 ohms. In regard to the switching frequency, because there is a range of the switching frequency that can vary in different embodiments (typically ranging from 18 kHz to 100 kHz), the impedance of the bypass capacitor at the high switching frequency can vary in proportion to the switching frequency. For example, at 80 kHz the impedance of the same 1 μF bypass capacitor would be 2 ohms. Typically, the impedance of the bypass capacitor at the operating switching frequency is typically less than 100 ohms.
Thus, the bypass capacitor does not substantially affect the rectified AC voltage provided via rectifier section 4 during operation of the ballast. The bypass capacitor present a high impedance to the rectified AC input which results in the AC current being distributed symmetrically on the rising and falling edges of the rectified AC voltage. In other words, the bypass capacitor causes the load current from the AC line to be sinusoidal to the load, thereby causing the load current to track the rectified AC voltage, which results in a high power factor. The tank circuit particularly, the inductor, is characterized to follow the rising and falling of the rectified AC voltage and thus present a sinusoidal current to the light source. The use of a high frequency, small value, non-electrolytic bypass capacitor is in distinction to the prior art that uses a low frequency, large value, electrolytic capacitor across the output of the rectifier to filter out the 120 Hz AC ripple due to the line frequency in order to remove the “valleys” in the rectifier output. The capacitance value of capacitor 102 in the embodiment of
The value of capacitor 102 can be around 0.22 μF for a 5 watt light source. The value can be adjusted as appropriate for the output load, but typically is 1 μF or less for a typical LED based light source that is less than 15 watts. The value of capacitor 102 is small enough so as to not impact the output rectified voltage at node 50. Specifically, the value should not preclude the output voltage presented at node 50 from dropping down to 30% to 15% or less of its peak voltage of the rectifier output at the end of each half cycle. In other words, the voltage at the bottom of the “valley” should be no more than 10-18 volts at 120 volts, and preferably lower. Thus, the bypass capacitor should not “smooth” out the rectified AC voltage.
One embodiment of the values of the components shown in
Those skilled in the art will realize that other values or type of components may be used, and that certain values may be modified for different sized loads or power supply voltages.
The output of the main portion 101 of the ballast (provided from the switching section) is identified as nodes 151 and 153. These nodes also serve as the inputs to the tank circuit 150, shown in
The resonant circuit can be viewed as a coupling device matching the impedance of the light source with the power source. The resonant circuit comprises an inductor 172 in series with a capacitor 170. The resonant circuit functions as an LC circuit that has a resonant frequency allowing energy to be alternately stored in the inductor and the capacitor. The resonant circuit can be characterized in one embodiment as generating an alternating voltage (e.g., a time varying voltage having a positive and negative value at different times). In addition, the resonant circuit can be characterized as providing an alternating current (e.g., a time varying current having a positive and negative value at different times). In many embodiments, a second capacitor may be added so as to provide an alternating current with a sinusoidal characteristic in the tank circuit. Thus, the resonant circuit may be viewed as a voltage source or current source, depending on how the load (e.g., rectifiers and LEDs) is coupled to the resonant circuit. In some embodiments, the coupling may occur using a transformer, which transforms the current/voltage on the primary winding (from the resonant circuit) to the secondary winding (to the rectifiers) according to well known principles.
The inductor 172 is generally a gapped core inductor that is capable of handling a peak current without fully saturating. The inductor processes both the lower line frequency current (e.g., 120 Hz) as well as the higher, switching frequency current (e.g., 20-100 kHz) and avoids saturation at the lower frequency. This is in contrast to prior art ballasts, which filter a rectified AC output voltage, resulting in a largely constant DC voltage with little 120 Hz ripple. Hence, the prior art inductors in the tank circuit (at least for gas-discharge light sources) are not designed to conduct a line frequency current because the ripple was removed by the smoothing capacitor. In
In one embodiment, the inductor can be a toroid shaped core about 1 to 1.5″ in diameter having about 90 turns of Litz wire providing for about one mH (milli Henry) or less of inductance. In one embodiment, the toroid is a Magnetics® Kool Mu® 007707A7 core. Such a toroid at 20 watts or less should be able to provide a high power factor (e.g., pf=0.9 or higher). As the power increases for the same size inductor, the power factor will decrease. While this power factor may be still higher than other ballast arrangements, it may drop below pf=0.9, and thus would not be considered a high power factor.
In other embodiments, the inductor can be a “double E” core with an air gap, or other configurations using a material with a distributed air gap. Other core configurations can be used as known by those skilled in the art. The load ratings of the ballast for LED lights sources are typically lower power compared to other types of light sources (e.g., gas discharge lamps), and hence the inductor can be relatively smaller in value and size.
Returning to
The presence of the inductor insures that when current flows into the resonant circuit when the upper switch closes, the current is in phase with the supply voltage, thereby contributing to the high power factor of the circuit. The inductor is also required for the resonant circuit to oscillate, thereby allowing energy to be transferred back and forth from the inductor to the capacitor. The resonant frequency of an LC circuit is described by equation 1 below:
where fR is the resonant frequency of the circuit, L is the inductance value of the inductor and C is the capacitance value of the capacitor 170.
The values of the inductor and capacitor components in the resonant circuit vary on the output power of the lamp and the desired resonant frequency. In Table 1 below, approximate values of the inductor and capacitor are indicated for certain embodiments, that are based on 120 VAC operation:
As evident, as the frequency increases, the inductor value decreases, allowing a smaller inductor to be used. This has an advantage in that it potentially allows a smaller size of the structure housing an integrated ballast and light source (“LED Bulb”). This may be desirable if the LED Bulb is intended as a replacement for incandescent bulbs. However, there is a practical upper limit of the switching frequency, because as the switching frequency increases, the overall system efficiency begins to decrease due to switching losses and other effects, such as the skin effect of the wire in the inductor.
The tank rectifier section 16 comprises in this embodiment a configuration 180 comprising four diodes 158a-158d. These are typically fast recovery diodes, such as 1NF4004 diodes, and are rated according to the current flow of the LED. In this embodiment which incorporates a single LED, the current requirements may be up to 1 or more amps. Since the voltage drop across the single LED light source is typically 3 volts, the current can be found according to Equation 2:
WattageLED/(VLED)=(CurrentLED). Eq. 2
Thus, a 6 watt LED with 3 volts across the LED would have 2 amps current flowing through it.
The LED light source section 18 comprises in this embodiment a single LED. In this embodiment, because the LED is in series with the resonant circuit, the current rating is typically between 20 ma-100 ma. However, as will be discussed below, in other embodiments of the tank circuit, other LEDs can be used that are capable of handling 1000 ma-2000 ma (1-2 amps) of current (or more) and which are available from various suppliers. Other high power LEDs, including those capable of handling up to 3-6 amps (or more), can be used as the light source. The LED is connected to the output of the diode rectifiers, and once the diode rectifiers exceed the threshold voltage required by the LED diode, current flows through the LED for generating light. Typically, in a single LED the forward voltage drop is about 3 volts.
The operation of the ballast can be described as follows. Household power comprising an AC voltage waveform is provided to the input of the input power section 2 and presented to the full wave bridge rectifier section 4. The AC waveform is transformed into a rectified waveform across nodes 50, 55. This waveform, shown as voltage waveform 200 of
The IC driver section and the transistor section cooperate to turn switch 144 (“upper switch”) and switch 145 (“lower switch”) alternately on and off. This occurs at a high frequency which is also referred to as the switching frequency. When the upper switch is closed, the voltage from node 50 (the time varying DC voltage) is provided to the tank circuit. When the lower switch is closed, the upper switch is open and no rectified line voltage is provided to the tank circuit. The resulting voltage waveform provided to the tank circuit is shown in
Thus, the input to the resonant circuit section comprises the square waves shown in
When the upper switch 144 closes, the voltage present at node 50 (which varies in value over time, as it is the rectified AC voltage), is provided to the resonant circuit. However, parasitic inductance in the power line to the ballast may inhibit current flow into the resonant circuit immediately after the upper switch closes. Thus, energy from the bypass capacitor discharges (because the capacitor will be at a higher potential) and provides current to the resonant circuit and ensures the resonant circuit continues operation. Then, as the inductance in the power line allows current from the power line to flow through the upper switch into the resonant circuit, further energy from the power line is provided into the tank circuit for the remainder of the half switching cycle. The charge from the bypass capacitor is relatively small, and is discharged within the half switching cycle. However, the bypass capacitor is sufficient in capacitor to ensure that current is flowing into the resonant circuit immediately after the upper switch closes. The bypass capacitor ensures the resonant circuit maintains resonance, and this is particularly applicable when dimming occurs, because no voltage is present from the rectified line voltage until the firing angle is encountered.
In the second half of the switching cycle, upper switch 144 opens and shortly thereafter, lower switch 148 closes. This essentially connects node 151 to node 153, which allows the energy in the resonant circuit to circulate therein. Essentially, energy is transferred between the inductor and capacitor, and current flow in the resonant circuit reverses direction. During the time when the lower switch is closed and the upper switch is open, the bypass capacitor 102 is being charged by the line voltage present on node 50. Consequently, when the next switching cycle begins, the bypass capacitor is charged and is ready to discharge when the upper switch closes, thus repeating the cycle. Thus, the current in the resonant circuit is continuously altering direction with energy continuously being introduced to maintain the cycle.
The value of the bypass capacitor must be sized within a range to achieve a desirable power factor and yet maintain operation of the resonant circuit. If, instead, the bypass capacitor were of such a large value (such as those prior art ballasts using an electrolytic smoothing filter capacitor), the bypass capacitor when discharging would provide so much current that the current drawn from the power source would be reduced. If the bypass capacitor were replaced with a smoothing capacitor that largely eliminated the voltage ripple in the rectified voltage, then current would be flowing into the tank circuit when the line voltage was crossing zero. This would result in current being drawn from the power source when the voltage was zero voltage. A large capacitor across the output of the rectifier would adversely affect the power factor of the ballast. Thus, the bypass capacitor is typically not an electrolytic capacitor and it is preferable to use a small value for the bypass capacitor such that a desirable power factor (e.g., from pf=0.7 or higher) is maintained during operation of the ballast. On the other hand, if the capacitor is too small, insufficient current would be provided to the ballast circuit when the upper switch initially closes, such that the resonant circuit may have insufficient current flow and ceases to function. Similarly, if the bypass capacitor is removed during operation, then the resonant circuit ceases to function.
This operation may be explained with the aid of
When the upper switch 310a closes, the lower switch 312a is open and the rectified voltage and current from the rectified voltage source 300 is allowed to pass through switch 310a into the resonant circuit. There typically is a slight delay in the current 305 from the power source 300 flowing into the ballast due to inductance in the wiring of the distribution system. Specifically, this includes inductance present in the distribution lines between the power source and the ballast. The wiring between the switch 310 and the commercial AC power source may include hundreds of feet of inside branch wiring in a building as well as wire outside the building, which has some small, but finite inductance. However, the inductance allows the current 305 to flow shortly after switch 310a closes. At the same time as switch 310a closes, bypass capacitor 314, which was previously charged, discharges 307 into the switch 310a, causing current 309, 311 to flow through the switch into the tank circuit. Thus, even if inductance in the power lines causes a momentary delay of the full current flow 305 from the power source, current 309 is flowing into the resonant circuit from the bypass capacitor to ensure that the resonant circuit maintains operation. The current provided by the bypass capacitor is relatively small in value, and quickly discharges, thereby providing high frequency energy to the circuit, so that it does not impact the flow of current 305 from the power source. The energy due to the current 309, 311 is stored in the resonant circuit 320 with some current flowing through the load, generating light. Note that this process occurs in the first half of the switching cycle. Thus, the charging/discharging of the bypass capacitor occurs many times during a cycle of the line voltage (e.g., 1/120 of a second) during which time the rectified input voltage is increasing and then decreasing. Thus, the current levels provided to the resonant circuit when the switch 310a closes vary, based on the level of the rectified AC voltage being switched. Because these current levels follow a sine wave in phase with the line voltage, a high power factor is achieved.
In
In the embodiment of
Recall that the switches operate continuously. Returning to
Returning to
However, when the voltage across the zener diode 114 exceeds a corresponding breakdown voltage (e.g., about −14.0 volts, etc.), the zener diode 114 enters what is commonly referred to as “avalanche breakdown mode” and allows current to flow from its cathode to its anode. In response, the current flows across the resistor 120 and causes the transistor 116 to have a base-emitter voltage (VBE), thereby turning ON transistor 116. The transistor 116 sinks current into the second node 55, which reduces the gate-source voltage of the NMOS transistor 110 and the current through the zener diode 114. Once the current in the zener diode 114 does not exceed the design of the output of the regulator value, the zener diode 114 recovers to the design value and reduces the current from flowing into the resistor 120. That is, by reducing the voltage at the source of the NMOS transistor 110, the voltage supplied to the driver circuit 132 does not substantially exceed the predetermined threshold voltage (Vmax). In the example of
Thus, the illustrated voltage regulator 6 is configured to provide a substantially constant (i.e., regulated) voltage to the driver 8. When the rectified voltage provided via the rectifier 4 falls below a predetermined threshold voltage (VT), the voltage output by the voltage regulator 6 decreases. However, the energy storage device 129 has a corresponding voltage that exceeds a minimum threshold voltage (VT) and continues to provide energy to the driver circuit 132. In addition, when the voltage at the node 50 falls below the voltage of the regulator 120, the diode 124 prevents current from flowing backwards from the capacitor 129 into the NMOS transistor 110 and resistor 108 from the constantly discharged tank circuit via 50.
Turning now to the resonant circuit, the current flowing into the resonant circuit at the line frequency is largely maintained as a sine wave, and is largely in phase with the voltage at the line frequency from the power source. Further, the resonant circuit does not store any significant energy (inductive or capacitive) to distort the low frequency current during the time period between the half cycles, thereby causing the resonant circuit to appear as a resistive load to the power supply. Thus, the present circuit maintains a high power factor during operation provided the inductor is sized appropriate. In other embodiments, the inductor may be sized smaller (so as to consume less physical space) but doing so reduces the power factor. Thus, it is preferable to size the inductor so as to obtain a power factor greater or equal to 0.7. In particular, because the current flowing through the resonant circuit is substantially similar to a sine wave, the crest factor of the illustrated example is approximately the square root of 2 (e.g., about 1.5), which is close to an ideal crest factor. Contrast this to the prior art ballasts which require a dedicated power factor correction circuit to obtain a suitable crest factor.
In the embodiment of
One alternative embodiment that can provide a higher current for a higher power LED while maintaining a lower current level in the resonant circuit is shown in
In the embodiment of
The resonant circuit can be modified as shown in
In the embodiment shown in
The voltage present across the primary winding of transformer 400 of
In this embodiment, for a 6 watt LED load, the bypass capacitor 102 may be 0.1° F., the resonant capacitor 170 may be 12 nF, the capacitor 402 may be 8.2 nF, the inductor may be 1 mH. These values are approximate. Further, because of variance in the tolerances of these parts, the switching frequency can be adjusted via the aforementioned potentiometer to tune the switching frequency to just above the actual resonant frequency. Adjustment of the potentiometer to adjust the switching frequency may be useful during manufacturing to compensate for component variances.
The wattage of a single LED has been traditionally limited by the materials used, and while new materials may allow greater power and light levels in a single LED, it is still desirable for many applications to have a light source producing more light than a single LED can produce. One solution is to use several lower power LEDs in series or parallel to generate more light. Further, these lower power LEDs are typically individually lower in cost. In one embodiment, a number of conventional LEDs are connected in series. These LEDs are typically conventional white-light emitting diodes, each having a 3 volt voltage drop. One such embodiment in shown in
In the embodiment shown in
To facilitate the voltage presented to the diodes 158 (which is the voltage at node 177, 179) and reaching the required voltage threshold, the tank circuit can be modified as shown in
In other embodiments, a series loaded configuration is also possible. In a series loaded configuration, the rectifier in the tank circuit generally relies, in some manner, on a sinusoidal current waveform in the resonant circuit in order to generate light in the LED. In such instances the voltage may be a square wave across certain elements. In a parallel loaded configuration, the rectifier in the tank circuit generally relies, in some manner, on a sinusoidal voltage in the resonant circuit in order to generate light in the LED. Typically in a parallel loaded configuration, the sinusoidal voltage is obtained across a first capacitor in the resonant circuit, which is configured as a voltage divider with a second capacitor in the resonant circuit. In some embodiments, the magnitude of the first capacitor across the primary of the transformer can exhibit aspects of both series and parallel loading configurations. As seen herein, transformers may be used in the tank circuit to modify the alternating current or alternating voltage characteristics in order to facilitate operation of the ballast.
Thus, until the voltage across capacitor 800 causes a current through the LED, there is no load offered by the LEDs 182 in the tank circuit. In other words, the load presented by the LEDs 182 is present only when the voltage across the capacitor 800 exceeds the required voltage drop. In summary, capacitor 800 ensures the voltage into the full wave bridge rapidly builds up rapidly allowing current to flow through the LEDs.
If there are few conventional LEDs connected in series, then capacitor 800 is less likely to be present. However, if there are a large number of LEDs connected in series, then capacitor 800 facilitates sufficient voltage to ensure there is current flowing through the LEDs and serves as the main capacitance of the tank circuit. Thus, various embodiments possible. The selection of how many LEDs can be driven is dependent on various factors, and the tank circuit can be modified to accommodate these options.
The benefit of combining the tank circuit section 150 with the main ballast section 101 is that it results in a high efficiency, high power factor, dimmable LED ballast that can be readily adapted for different LED configurations. In order to accomplish this, the inductor should be sized so as to maintain operation in the non-saturated mode.
Another tank circuit embodiment is shown in
The transformer 1110 in this case has a center tapped secondary winding. Thus, the secondary has three outputs 1115a, 1115b, and 1115c. The center tap 1115b is connected to the cathode of the LED 182, and each of the outer secondary winding connections 1115a, 1115b are connected to the anode of the LED via a respective diode 1120, 1122. During operation, namely during a first part of the switching cycle, the LED 182 is receiving current from the upper secondary winding, namely connection 1115a, with current passing through diode 1120 through the LED 182, and back to the center tap 1115b. During the other half of the switching cycle, current is flowing from other connection 1115c through the diode 1122, to the anode of the LED 182, and back to the center tap secondary winding, connection 1115b. In this embodiment, during each cycle, there is only one diode for which there is a rectifier diode voltage drop. Other variations on
This embodiment only involves two rectifying diodes 1120 and 1122, so that a lower diode voltage drop represents greater efficiency of operation compared to using four diodes. Thus, this improves the rectification efficiency by 100% relative to using a full bridge rectifier configuration. For a ballast using only a single LED, the reduced voltage drop in the rectifying section 16 represents a significant increase in efficiency, relative to using four diodes. Although this embodiment can also be used with multiple LEDs, the relative efficiency gains are not as great as the number of LEDs increases.
The embodiment of
During operation, the voltage from the main windings (e.g., terminals 1242 and 1244) provide a voltage that is rectified by MOSFET 1224a and 1224b respectively. The gate control windings provide a voltage greater than that generated by the main windings. The MOSFETS are turned ON when the voltage at terminal 1241 increases above a threshold amount above the voltage at node 1242 thereby allowing the gate to turn the MOSFET ON. The resistor 1220a and zener diode 1222a limit the current and voltage so that the MOSFET is only turned ON at the appropriate times in a synchronous manner. Similarly, the corresponding components for MOSFET 1224b turn ON at complimentary times. In this manner, the time varying DC voltage generated from the resonant alternating voltage in the resonant circuit is provided to the LED to produce light.
Another embodiment of the tank circuit is illustrated in
During operation, the current from each inductor is added to provide the current through the LED, but the secondary winding only carries half of the output current (hence, the name “current doubler”). The rectifiers 1120, 1122 function as described previously, and in other embodiments, these diodes may be part of a MOFSET to provide further efficiency gains.
Still another embodiment is shown in
The various embodiments of the ballast can be effectively dimmed using a conventional triac based phase control dimmer, including the dimmer disclosed in U.S. patent application Ser. No. 12/205,564 filed on Sep. 5, 2008, which in turn claims the benefit under 35 U.S.C. § 119(e) to U.S. Provisional Patent Application entitled “Two-Wire Dimmer Switch for Dimmable Fluorescent Lights” filed on Feb. 8, 2008, bearing Ser. No. 61/006,967, both of which are herein incorporated by reference for all that each teaches (referred to as “Two Wire Dimmer”).
The effect of the Two Wire Dimmer on the incoming supply voltage to the ballast is shown in
In
Although there is no voltage to the ballast provided during time period 1002a, there is sufficient voltage provided to the IC driver, allowing the switching of the switches to continue during this time period. Recall that there is a housekeeping capacitor in a voltage regulator that provides power to the IC, and the charging of the housekeeping electrolytic capacitor in the voltage regulator is performed at the very beginning of the voltage waveform produced from the output from the dimmer. The charging of the housekeeping capacitor dissipates the stored inductance in the house wiring that is created when the phase controlled dimmer is turned ON. This would normally cause a ringing of current of the input bypass capacitor if it were not damped by the load presented by the series regulator at this precise time during the charging of the housekeeping capacitor. The housekeeping capacitor also provides energy to the IC driver during the portion 1002a when there would otherwise be insufficient voltage to driver the IC.
Further, the ballast may also incorporate an optional resistor 103 in the power input section (see
The impact of dimming on the voltage output of the switching power is shown in
During the portion 1002b, there is no voltage provided to the ballast. The switching of the switches continues during this time period, so that when the upper switch closes, there is no energy provided to the ballast. While the bypass capacitor provides some charge to the resonant circuit when the upper switch closes, the bypass capacitor by itself does not have enough charge to maintain operation of the resonant circuit during period 1002b. The absence of any energy into the tank circuit causes the energy in the resonant circuit to quickly reduce. Once the voltage available to the diode rectifiers in the tank circuit drops below a certain voltage, no further current will be drawn from the resonant circuit and no light is generated. However, even though the energy in the tank circuit reduces to a level that is not able to generate light the LED, the resonant circuit is still resonating, albeit at a diminishing energy level with each switching cycle.
When the phase dimmer restores the rectified line voltage at 1061, energy is provided back into the tank circuit via the upper switch. Because the switches continuously operate in synchronization with the resonant circuit, the energy level can be quickly restored and light is quickly regenerated by the LED. Because the voltage at the rectified line voltage appears as a “step function” at point 1061, a high level of voltage is provided to the tank circuit to immediately energy it. However, the existence of the zero-voltage portion 1002b reduces the average current available to the LED light source during a half cycle of the line frequency, and thus, the average light generated must also be reduced.
Further, in the case of dimming, during period 1002b, there is no voltage, and hence no current drawn by the resonant circuit. This reduces the overall energy consumed by the ballast overall. The power factor during operation with a dimmer is slightly reduced relative to operation without it. However, for the portion of rectified line voltage that is non-zero, the current draw of the ballast is largely in phase with the line voltage. Consequently, even with dimming, the power factor is relatively high.
Although certain methods, apparatus, systems, and articles of manufacture have been described herein, the scope of coverage of this patent is not limited thereto. To the contrary, this patent covers all methods, apparatus, systems, and articles of manufacture fairly falling within the scope of the appended claims either literally or under the doctrine of equivalents.
This application is a continuation-in-part of U.S. patent application Ser. No. 12/277,014 filed on Nov. 24, 2008, which is a continuation-in-part of U.S. patent application Ser. No. 12/187,139 filed Aug. 6, 2008, which is a continuation-in-part of U.S. patent application Ser. No. 12/178,397 filed on Jul. 23, 2008, which in turn claims the benefit under 35 U.S.C. § 119(e) to U.S. (Provisional) Patent Application entitled “Dimmable Ballast with High Power Factor” filed on Feb. 8, 2008, Ser. No. 61/006,965, the contents of which are herein incorporated by reference for all that each teaches.
| Number | Date | Country | |
|---|---|---|---|
| 61006965 | Feb 2008 | US |
| Number | Date | Country | |
|---|---|---|---|
| Parent | 12277014 | Nov 2008 | US |
| Child | 12537464 | US | |
| Parent | 12187139 | Aug 2008 | US |
| Child | 12277014 | US | |
| Parent | 12178397 | Jul 2008 | US |
| Child | 12187139 | US |
