This invention generally relates to AC-to-DC power converters. In particular, this invention relates to LED drivers.
Light emitting diode (LED) lighting is a fast growing industry due to the high efficiency and long life of LEDs. One difficulty of using LEDs stems from the large mismatch between the alternating current (AC) mains voltage, typically in the range of 100 VAC-277 VAC and the voltage of a single LED which is typically on the order of 1-2V. Another difficulty stems from the range of LED voltages as a function of temperature, manufacturer tolerances, and different manufacturer specifications. Still, another difficulty stems from the fact that LEDs are (direct current) DC devices whereas the primary source of power is AC.
The LED voltage mismatch may be reduced by using long series strings of LEDs. However, this only alleviates part of the issue since it is typically not feasible to place so many LEDs in series to match the AC mains voltage. Furthermore, placing devices in series only partly addresses the issue of voltage matching and does not address the issue of AC-to-DC mismatch or LED voltage variation.
A simple, low-cost solution is to place a large value resistor and a high-voltage diode in series with the LED string. However, this solution is very inefficient, has lifetime issues due to the heating of the resistor, and also leads to a very poor utilization of the available LED power due to the extremely high ripple current produced by the LED.
Many AC-to-DC drivers have been proposed and brought to market to address the issues of driving an LED. One such driver is discussed in U.S. Pat. No. 6,304,464 which proposes a flyback converter as an LED driver and represents the power conversion method used in the majority of AC-to-DC LED drivers which are on the market. While this typical type of driver provides a DC voltage to the LED, these driver types suffer from several drawbacks. One drawback of these drivers is the use of limited-lifetime components which gives the driver a much lower effective lifetime than the LED itself. The limited lifetime components include electrolytic capacitors used as the main storage element and optocouplers used in the feedback loop. These low-lifetime components not only reduce the cost-effectiveness of the overall LED solution, but they also limit the applications to use over relatively small temperature variations. A further drawback of these LED drivers is their inability to provide a lighting solution which provides a specific light level across temperature and manufacturing tolerance variations. Typically, LED drivers regulate the voltage across the LED string. The current is therefore determined by the forward voltage drop of the LEDs and the resistance of the LEDs. Small changes in LED voltage can lead to a large change in LED current and consequently to a large change in light output.
High-power drivers, such as those above 75 W in power, usually incorporate power factor correction on the input. Standard power factor correction circuits use either fixed-frequency continuous-conduction-mode pulse-width-modulation or variable-frequency critical-conduction-mode pulse-width-modulation. Fixed-frequency continuous-conduction-mode pulse-width-modulation typically requires expensive controllers, very large inductors, and large EMI filtering components to reduce the noise created at the single pulse-width-modulation frequency. Furthermore, fixed-frequency controllers can have high switching losses since the frequency is held constant regardless of the waveform amplitude. On the other hand, variable-frequency critical-conduction-mode pulse-width-modulation is inefficient due to the very high ripple current produced in the inductor, and therefore also requires large filters to reduce electro-magnetic-interference (EMI).
Traditional converters use an electrolytic storage capacitor for several reasons including the following: 1) Electrolytic capacitors are relatively inexpensive compared to most other types of capacitors for a given value of the product of capacitance and voltage rating. 2) The large capacitance of electrolytic capacitors allows significant reduction of ripple voltage and can therefore be used to provide a relatively constant output voltage. 3) The small size of electrolytic capacitors provides the ability to make relatively small drivers.
The prior art converter illustrated in
While this prior art converter in
Specifically power-factor-correction converters are typically operated in one of two basic control methodologies. The first basic control methodology is referred to herein as critical conduction mode, in which the current through switch S201 is ramped up to a current proportional to the input voltage, and then commutated to D205 when the semiconductor switch is turned off. When the current through L201 decays to zero, switch S201 is then turned on again. The net result is an average current through L201 which is proportional to the input voltage. The frequency varies throughout the ac grid cycle. A great drawback to this control method is that the peak-to-peak ripple current through L201 is always twice as large as the instantaneous current that is drawn from the ac grid. Thus, L201 must be designed to saturate at nearly double the value of current at which it would otherwise be designed, there are large losses due to the high ripple current, and the AC EMI filter must be designed to filter out very large differential currents. This method is typically used for relatively low power power-factor-correction converters less than approximately 120 W due to the cost savings that occur from using a diode D205 which may have some recovery losses.
The second basic control methodology is referred to herein as continuous conduction mode. In this method of operation, switch S201 is operated at constant frequency pulse-width-modulation. However, the duty cycle is controlled to cause the current through L201 to be primarily sinusoidal in phase with the AC grid voltage. Some drawbacks to this method of control include the following: relative complexity of the control compared with the critical conduction mode method, similar ripple amplitude near the zero-crossings of the AC grid current compared with the peak of the grid current, thus causing increased harmonic distortion, and substantial EMI noise concentrated at multiples of the pulse-width-modulation frequency.
Embodiments of the present invention solve the above-mentioned problems and provide a distinct advance in the art of LED drivers. One embodiment of the invention provides an LED driver with an AC power source coupled to a first magnetic component with an inductance, which is further coupled to a first controllable semiconductor switch and to a DC bus comprising a film capacitor. The DC bus is further coupled to a string of LEDs, also referred to herein as an LED load. A first controller controls the first semiconductor switch in such a way as to draw a sinusoidal current from the AC power source and such that the film capacitor absorbs pulsating power from the power source and provides DC power to the LED string.
Embodiments of the present invention have the advantage of using only non-electrolytic storage elements and non-optical feedback components to provide a high lifetime product that can match and even exceed the lifetime of the LEDs. In an embodiment of the present invention, the film capacitor is sized such that the peak-to-peak AC ripple power in the LED load is greater than 20% of the steady-state power in the LED load.
In another embodiment of the present invention, the LED driver further comprises a non-regulated isolated DC-to-DC converter that functions as a DC transformer and is coupled to the DC bus and to the string of LEDs. In still another embodiment of the present invention, the LED driver further comprises a first controller that produces a first signal and a second signal. The first signal and second signal are rectified sinusoids with a DC offset and are in phase with each other such that the amplitude of the first signal is less than or equal to the amplitude of said second signal, and the sinusoidal portion of the second signal divided by the sinusoidal portion of the first signal is a constant over the course of each half-cycle of the ac power source.
Furthermore, the first controller compares the current flowing in the first magnetic component to the first signal and the second signal to determine whether to turn on the first controllable semiconductor switch in such a way as to either decrease or increase the current through the first magnetic component and in such a way as to produce a varying pulse-width-modulation frequency which decreases as the instantaneous value of the current increases, and which produces a value of AC ripple current which is smaller than the instantaneous value of the AC current. This advantageously allows use of an inexpensive controller, allows the user to easily trade switching losses for input current total harmonic distortion, and provides an easy method of control to provide a spread-spectrum EMI signature, thus reducing EMI signature at any specific frequency.
The LED driver may also adjust a first predetermined current level of an LED string as a function of LED voltage in such a way as to cause the power in the LED string to remain constant when the LED string voltage changes. This adjustment can be done, for example, by linearly reducing the predetermined current level with increasing LED string voltage. Furthermore, in some embodiments of the invention, the single-AC-power-cycle average value of inductance of the first magnetic component changes with load such that the average inductance value when operating at full load is less than 70% of the average inductance value when operating at 10% load. This variable inductance value may be enabled through a stepped air gap in the core of the first magnetic component.
In another embodiment of the invention, the controller may employ a multiplier which multiplies a reference sinusoidal signal by a multiplicand, such that the multiplicand changes at a slow rate compared with the frequency of the input power source and the multiplicand is increased when the current in the LED string is below the first predetermined current level, and the multiplicand is decreased when the current in the LED string is above the first predetermined current level. Furthermore, a first signal providing information about the comparison between the first predetermined current level and the LED string current is transmitted across a high-voltage isolation boundary using a first transformer, and the voltage at the LED side of the DC-to-DC power transformer is gated to produce the first signal.
This summary is provided to introduce a selection of concepts in a simplified form that are further described below in the detailed description. This summary is not intended to identify key features or essential features of the claimed subject matter, nor is it intended to be used to limit the scope of the claimed subject matter. Other aspects and advantages of the current invention will be apparent from the following detailed description of the embodiments and the accompanying drawing figures.
Embodiments of the current invention are described in detail below with reference to the attached drawing figures, wherein:
The drawing figures do not limit the current invention to the specific embodiments disclosed and described herein. The drawings are not necessarily to scale, emphasis instead being placed upon clearly illustrating the principles of the invention.
The following detailed description of the invention references the accompanying drawings that illustrate specific embodiments in which the invention can be practiced. The embodiments are intended to describe aspects of the invention in sufficient detail to enable those skilled in the art to practice the invention. Other embodiments can be utilized and changes can be made without departing from the scope of the current invention. The following detailed description is, therefore, not to be taken in a limiting sense. The scope of the current invention is defined only by the appended claims, along with the full scope of equivalents to which such claims are entitled.
In this description, references to “one embodiment”, “an embodiment”, or “embodiments” mean that the feature or features being referred to are included in at least one embodiment of the technology. Separate references to “one embodiment”, “an embodiment”, or “embodiments” in this description do not necessarily refer to the same embodiment and are also not mutually exclusive unless so stated and/or except as will be readily apparent to those skilled in the art from the description. For example, a feature, structure, act, etc. described in one embodiment may also be included in other embodiments, but is not necessarily included. Thus, the current technology can include a variety of combinations and/or integrations of the embodiments described herein.
A light emitting diode (LED) driver 10, constructed in accordance with embodiments of the present invention, is shown in
The LED driver 10, as illustrated in
The AC EMI filter 310 may be configured to reduce high-frequency current from reaching the utility grid 100. The input rectifier 320, receiving input from the AC EMI filter 310, as illustrated in
As illustrated in
Unlike most prior art LED drivers that encompass power conversion and use electrolytic capacitors for energy storage, the LED driver 10 of the present invention uses a film capacitor. The use of film capacitor C301, as illustrated in
The above points are illustrated by the curves on the chart in
During normal driver operation, the LLC controller 371 produces fixed duty cycle, fixed frequency gate pulses to switch S302 and switch S303 such that the gate drive pulses of switch S302 and switch S303 are phase shifted 180 degrees from each other. The duty cycle of the pulses is 50% minus a small time period needed for the current in the switches to commutate to the opposing switch. For example, a typical duty cycle would be 48%. The switching frequency of the LLC converter 360 is tuned to operate at frequencies slightly below the natural resonant frequency of the leakage inductance of transformer TX301 and the capacitance of capacitor C302. As a result of operating at resonance, the impedance of capacitor C302 is cancelled by the impedance of the leakage inductance of TX302 and the output voltage of the LLC converter 360 is very tightly coupled to the input voltage of the LLC converter 360. The LLC converter 360 therefore acts as a DC transformer with a turns ratio equal to one-half of the turns ratio of transformer TX301. (The factor of one-half is produced by use of a half-bridge rather than a full-bridge). The LLC converter 360 operates with zero-current switching and close to zero-voltage switching. It therefore operates at very high efficiency (such as 98%). The LLC converter 360 does not regulate the output voltage since the output voltage is always a scaled multiple of the input voltage for the LLC converter 360. The LLC converter 360 provides high-voltage isolation between the LED string and the AC grid 100 and also provides voltage scaling appropriate for the load 350 that is being used. The ratio of double-AC-grid frequency ripple voltage to DC voltage will therefore be the same at the input and the output of the LLC converter 360.
As illustrated in
Other non-regulated isolated converters may be used in place of an LLC converter to perform the same functions described herein. For example, a hard-switched half-bridge that is operated at 50% duty cycle and followed by a transformer will perform a similar function. Furthermore, full-bridge versions of these converters perform the same function. Other possibilities will occur to those skilled in the art. What is important is that this converter stage be optimized for high-efficiency design and designed to act as a DC transformer.
LED controller 374 may be configured to monitor the current in the LED load 350 and to send a signal to isolator 375, which gates a pulse-width-modulated signal to multiplier 373 depending on whether the measured current is below or above a predetermined level of current. The predetermined level of current can be easily altered with a dimming signal to provide a dimming function for the LED driver 10. Furthermore, some embodiments of the current invention adjust the predetermined level of current as a function of voltage across the LED load 350 in such a way as to regulate the power into the LED load 350 to a nearly constant level. One low-cost method for producing a nearly constant LED power regardless of LED voltage is to linearly decrease the predetermined LED current as the LED voltage is increased.
Referring again to
In the embodiment of the invention illustrated in
The power-factor-correction controller 372 uses the output of the multiplier 373 with two scaling factors and two offset factors to provide upper and lower boundaries for the power-factor-correction current. As illustrated in
There are several benefits to the operation of the power-factor-correction controller 372 compared to standard methods of operation of power-factor-correction controllers including the following: 1) Provided that k2 is larger than k1 (which would be the recommended way to operate the converter), the frequency will be lower and the ripple will be higher at the peak of the ac grid than at the zero-crossings. This causes lower losses and lower total harmonic distortion than continuous-conduction-mode constant frequency operation. 2) The ripple is significantly lower than the instantaneous value of the ac grid current. This means that losses and total harmonic distortion will be much lower than would be the case for critical-conduction-mode operation. 3) The total harmonic distortion and losses can easily be traded by adjusting the k1, k2, V1, and V2. Whereas with critical-conduction-mode operation, no parameters can be controlled except through inductance value and continuous-conduction-mode only allows control of the constant switching frequency. 4) The frequency is lowest when the amplitude of the current is highest. Thus the EMI generated is lower than for the continuous-conduction-mode method which has a constant frequency. Also, since the amplitude of the current is significantly lower than for the critical-conduction-mode method, the EMI generated is also significantly lower than for the critical-conduction-mode method. The proposed method of controlling the power-factor-correction converter is therefore advantageous compared with traditional methods of control in regards to efficiency, total harmonic distortion, EMI, and ability to easily trade off total harmonic distortion with efficiency.
The LED driver 10 illustrated in
Further efficiency and cost benefits can be realized by designing the inductance of L301 to significantly vary with load. For example, L301 can be designed to decrease in inductance to only 70% of its value or less when the load increases from 10% load to full load. The increase in inductance at small loads will also cause the ripple at the zero-crossings of the AC grid cycle to decrease compared with the ripple at the grid peaks, thus reducing total harmonic distortion and reducing switching frequency near the zero-crossings. The decrease in switching frequency near the zero-crossings will also decrease the losses.
In practice, there are many known methods of designing an inductor (e.g., L301 in
In another alternative embodiment of the present invention, two or more DC-to-DC converters (such as the LLC converter 360 described above) may be coupled to the film capacitor C301. For example, each DC-to-DC converter transformer may be matched to the specific LED string that needs to be driven by that transformer. Furthermore, in this alternative embodiment of the invention, the LED controller 374 may be duplicated for each DC-to-DC converter. The function of the switch S304 may then be changed to an “AND” function from all of the DC-to-DC converters. That is, if any of the LED strings reaches or exceeds their corresponding predetermined level of current, the pulse-width-modulation signal input of multiplier 373 may be disabled, whereas if none of the LED strings exceed their corresponding predetermined level of current, the pulse-width-modulation signal may be enabled.
In an alternative embodiment of the multiplier (not shown), the pulse-width-modulation signal on the secondary can be gated to charge or discharge a capacitor on the primary side of the circuit. The capacitor voltage can then be multiplied by the sinusoidal reference signal voltage through use of a junction gate field-effect transistor (JFET) or other multiplier.
Advantageously, the LED driver 10 described herein can make use of a non-electrolytic capacitor as its main storage element (so that it can have a long lifetime and high reliability operating at high temperatures for outdoor applications). Furthermore, the LED driver 10 can operate at high efficiency and may have an inexpensive feedback loop that does not use optical components. Of further benefit is the power-factor-correction controller 372 described herein, which reduces harmonic distortion, spreads the EMI noise across many frequencies, and allows use of an inexpensive controller.
Although the invention has been described with reference to the embodiments illustrated in the attached drawing figures, it is noted that equivalents may be employed and substitutions made herein without departing from the scope of the invention as recited in the claims.
Having thus described various embodiments of the invention, what is claimed as new and desired to be protected by Letters Patent includes the following:
This non-provisional application claims priority to U.S. Provisional Patent Application Ser. No. 61/924,101 filed on Jan. 6, 2014, titled “LED Driver,” which is herein incorporated by reference in its entirety. This application and the Provisional patent application have at least one common inventor.
Number | Date | Country | |
---|---|---|---|
61924101 | Jan 2014 | US |