A matter of light, Part 1---The ABC's of LEDs

Conserving energy is now a mandate, not a choice, and part of that mandate is the need to go Green. When it comes to lighting, we can easily imagine the impact of globally improving the efficiency of lighting sources by 10 percent. But what if it could be improved by 1000 percent? Newly enhanced light emitting diodes (LEDs) have the potential to achieve these efficiency improvements while maintaining high performance and reliability that supersede many currently used sources. In the first part of this four-part series, we look at the LED's physical structure, range of colors, efficiency, LED drivers, and applications. Anatomy Physically, LEDs resemble p-n junction diodes. As with p-n junctions, electrons and holes flow towards the junction when a positive differential voltage is applied between the anode (p-side), and cathode (n-side). Once an electron is recombined with a hole, it releases energy. Depending on the physical properties of the p-n junction materials, the released energy can be non-radiative, as for the typical diode applied in discrete circuits, or in the form of emissions in the optical range. For an LED, the wavelength of the emitted light (i.e., its color) depends on the band gap characteristics of its p-n junction material. Performance-wise, LED materials have relatively low reverse breakdown voltages since they have relatively low band gaps. Colors Red LEDs were the first to become commercially available in the late 1960s but their light output was very low. Despite this shortcoming, they were commonly used in seven-segment displays. Thanks to advancements in material science, nowadays LEDs are commercially available in a variety of colors with some of them having light outputs that would blind you if you stared directly at them. Blue LEDs became widely available a few years ago. Mixing blue LEDs with red and green LEDs produces white light. This technique of generating white light provides a large color gamut, dynamic light tuning, and excellent color rendering (CRI), which is well suited for high-end backlighting applications. A simpler and more economical way of producing white light is to use blue LEDs and a phosphor coating that converts some of the blue light to yellow. The yellow light stimulates the red and green receptors of the eye, and therefore mixing blue and yellow gives the appearance of white. This scheme can provide good CRI but the LED's light output may suffer from inconsistent color temperatures due to manufacturing discrepancies and varying thicknesses in the phosphor coating layer. (Click on Image to Enlarge)

Figure 1: LED color chart for the basic colors

Efficiency High efficiency is the buzz word for LED-based light sources. When it comes to lighting, efficiency is defined as the light output per unit power. Thus, in the metric system, it is measured in lumens (lm) per watt (W). Recently some LED manufactures introduced LEDs with promised efficiencies hitting the 150 lm/W mark. In comparison, incandescent comes in at 15 lm/W, and fluorescent provides 70 lm/W. So could LEDs put incandescent and fluorescent out of business any time soon? Maybe, but, unfortunately some of these LED's efficiency numbers are subject to specmanship. The problem is that LED inefficiency has to do more with the fact that a considerable portion of the produced light is reflected at the surface of the packing material back into the LED die. This reflected light is likely to be absorbed by the semiconductor material and turned into heat. Anti-reflection coating, and minimizing the reflection angles by using a half-sphere package with the LED placed at the center, reduces the amount of reflected light and improves efficiency. However, these techniques are subject to manufacturing variations and may require high premiums to ensure consistent performance. So while LEDs are rapidly being adopted by industry, they've got a long way to go. Applications There are many factors which make LEDs eye-catching for high-performance modern electronics. For example, their higher light output per watt extends battery life and thus they are well suited for portable applications. In addition, an LED's fast turn-on/turn-off characteristics fit perfectly with the needs of automotive tail lights, especially the brake lights, since it improves safety by providing drivers more response time. RGB LEDs in backlighting complies with ROHS standards, since LEDs do not contain lead or mercury. LED lighting facilitates a full-spectrum light source with larger color gamut. LEDs have an exceptionally long lifespan, which enables their use in applications where long-term reliability is highly desirable, such as traffic lights. Machine vision systems require a focused, bright, and homogeneous light sourceLEDs are a great match. LEDs, with their simple-to-implement dynamic light-tuning, would also allow you to set the light in your living room to green when you need to relax and to red when it's time for bullfighting. Drivers LEDs are inherently current-driven devices; i.e., their brightness varies with their forward current, IF. Depending on the color as well as the forward current, the LEDs' forward voltage drop, VF, varies as well. Thus, driving LEDs with a constant current is essential to achieve the desired color and brightness level. An LED driver scheme can be as simple as a voltage source and a ballast resistor (Figure 2a). This solution works best for narrow-input range, lowcurrent applications in which the LED's forward voltage drop is slightly below the input supply voltage. But variations in the input supply voltage or the LED forward voltage drop will increase the LED current and, therefore, the light intensity and the color will shift. (Click on Image to Enlarge)

Figure 2: Simplified LED driver schemes Linear regulators can be used to provide tighter LED current control in small step-down ratio applications (Figure 2b). In the case of low-current step-up requirements, switching capacitor circuits can be utilized (Figure 3c). For wide-input range, high-current applications, simple driver schemes such as those mentioned above unfortunately yield high power dissipation and poor efficiency. Consequently, more efficient and relatively more complex solutions such as switching regulators are required (Figure 2d). Switching regulators process power by interrupting the power flow and controlling the conversion duty cycle, which results in pulsating current and voltage. They can be configured in isolated and non-isolated configurations to realize voltage or current step-down (buck), step-up (boost) or both (buck-boost) functions. In general, designers select a switching-regulator topology based on a tradeoff between cost and desired performance at a given power conversion requirement. On the other hand, in order to properly drive LEDs, the switching regulators should be configured as constant-current sources. Which switching-regulator topology can be simply configured as a current source while providing an optimal tradeoff between cost and performance? Stay tuned for part two of this series. About the authors Sameh Sarhan is a staff applications engineer for the Medium Voltage/High Voltage Power Management group in Santa Clara, CA. He has been involved with power electronics in various forms since 1998, having worked for FRC Corp. and Vicor Corp. His experience includes the design of hard/soft switching power supplies from a few watts to 600 watts. Sameh received a bachelor's degree in electronics engineering in 1996 from Cairo University (Egypt). Chris Richardson is an applications engineer in the Power Management Products group, Medium and High Voltage Division. His responsibilities are divided between lab work, bench evaluation of new ICs, written work such as datasheets and applications notes, and training for field engineers and seminars. Since joining National Semiconductor in 2001, Chris has worked mainly on synchronous buck controllers and regulators. In the last three years he has focused on products for the emerging high brightness LED market in the automotive and industrial areas. Chris holds a BSEE from the Virginia Polytechnic Institute and State University.

Design Article
A matter of light, Part 2--- Buck whenever possible
Sameh Sarhan and Chris Richardson, National Semiconductor 5/27/2008 9:22 AM EDT

In part one of this series, we thrashed out the basics of LED lighting sources and their driving requirements. The performance of simple driving techniques, such as voltage sources/ballast resistors and linear regulators, fall short as the complexity and input power requirements of LED-based lighting sources increase. Thus a more sophisticated switch-mode LED driver is required. So what would be the topology of choice? In part 2, we discuss why a constantcurrent buck converter should be the first preference when it comes to switch-mode LED drivers or, in other words, why the buck should be used whenever possible. The rapid adoption of LEDs in various applications makes simple drive solutions such as linear regulators impractical in many cases. In general, simple drive schemes continuously deliver power from the input source to the driver's output while using resistive elements to program the desired LED forward current. For the same LED current, the losses in these resistive elements increase considerably as the line voltages increase. For example, a linear regulator based LED driver yields 70 percent efficiency when supplying 1 amp from a 5-volt input source to a typical white InGaN LED (VF = 3.5V). Under the same operating conditions, the driver's efficiency will drop to approximately 30 percent when the input voltage increases to 12 volts. Such poor efficiencies require impractical thermal management schemes. Switching regulators Switching regulators improve the conversion efficiency. They interrupt the power flow while controlling the conversion duty cycle to program the desired output voltage or output current. Interrupting the power flow results in pulsating current and voltage and therefore it necessitates the use of energy storage elements (inductors and/or capacitors) to filter these pulsating waveforms. Contrary to linear regulators, switching regulators can be configured in different arrangements to realize voltage or current step-down (buck), step-up (boost) or both (buck-boost) functions. They are also capable of achieving high conversion efficiencies across wide input/output range. Replacing the linear regulator with a buck-based LED driver in the previous example yields 95 to 98 percent efficiency across the 5-to-12 volt input range. The configuration flexibility and the efficiency improvements of switching regulators come at the expense of higher noise generation caused by the periodic switching events, as well as higher premiums and reduced reliability due to their perceived complexity. Constant-current LEDs favor regulator topologies that can be simply configured as a constant-current source. The selected topology should also combine high performance with minimum componentcount to increase the driver's reliability and to reduce cost. It should also facilitate the use of various dimming techniques to take advantage of the LEDs dynamic light- tuning characteristic. Fortunately, the most basic step-down (buck) switching topology enjoys all these characteristics, making it the regulator of choice to drive LEDs whenever possible. Constant-current power stage Switching regulators are most commonly known as voltage regulators. Figure 1a illustrates a basic constant-voltage buck regulator. The buck controller maintains a constant output voltage as the line voltage changes by varying the operating duty cycle (D) or the switching frequency. The desired output voltage set point is programmed using the following equation (Eq. 1):

Fig. 1a: Basic step-down (buck) voltage regulator The inductor, L, is selected to set the peak-to-peak current ripple, Ipp, while the capacitor, Co, is selected to program a desired output-voltage ripple and to provide output-voltage hold-up under load transients. The average inductor current in a buck converter is equal to the load current, and, therefore, we can set the load current by controlling the peak-to-peak inductor-current ripple. This significantly simplifies the conversion of a constant-voltage source into a constant-current source. Figure 1b illustrates a basic constant-current buck regulator. Similarly, constant-current buck regulators provide line regulation by adjusting the conversion duty cycle or the switching frequency, and the LED current, IF, is programmed using the following equation (Eq. 2):

Fig. 1b: Basic step-down (buck) current regulator After we set the LED current, IF, we must properly sense the inductor current. Theoretically, multiple current sense schemes such as MOSFET Rdson sensing and inductor DCR sensing can be used. However, practically, the current sense precision of some of these would not meet the required LED current set point accuracy (5 to 15 percent for a high brightness LED (HB-LED). If we directly sense IF through an inline resistor, RFB, we can secure the needed

precision, but there may be excessive power dissipation in the current-sense resistor. Lowering the feedback voltage, VFB, allows the use of lower resistance values for the same IF (Eq. 2), which minimizes losses. The newer dedicated LED drivers generally offer reference voltages (feedback voltages) within the range of 50 to 200 millivolts.

Uniquely, constant-current buck-driven regulators can be configured without output capacitance. The use of the output capacitor, Co, in these regulators is limited to AC current filtering since they inherently do not experience load transients and have continuous output currents. When we configure a constant-current buck regulator without output capacitance, we substantially increase the converter's output impedance and, in turn, boost the converter's ability to rapidly change its output voltage so that it can maintain a constant current. As a result, the dimming speed and dimming range of the converter improve significantly. Wide dimming range is valuable feature in applications such as backlighting and machine vision. On the other hand, lacking the required output capacitance, AC-current-ripple filtering circuitry necessitates the use of higher inductance values in order to meet the LED manufacturers recommended ripple current (IF = 5 to 20 percent of the DC forward current). At the same current rating, higher inductance values increase the size and cost of the LED driver. Consequently, the use of output capacitors in constant-current buck-based LED drivers is governed by a tradeoff between cost and size versus dimming speed and dimming range. For example, in order to drive a single white LED (VF 3.5 volts) at 1 amp with a ripple current, IF, of 5 percent from an input of 12 volts at 500 kHz requires a 50 microhenry inductor with a current rating of 1.1 amps. However if the inductor ripple-current is allowed to increase to 30 percent, then the inductance required is less than 10 microhenries. For the same core material and at approximately the same current rating, a 10 microhenry inductor will be typically offered at roughly half the size and cost of a 50 microhenry inductor. To attain the desired IF (5 percent) using the 10 microhenry inductor, the output capacitance required is calculated based on the dynamic resistance, rD, of the LED, the sense resistance, RFB, and the impedance of the capacitor at the switching frequency, using the following expression (Eq. 3):

where (Eq. 4):

Control-loop schemes Buck-based power stages are well-matched to several control-loop schemes and free of stability limitations such as right-half-plane zeros. They uniquely facilitate the shunt PWM dimming approach in addition to being compatible with other dimming methods. This provides the system designer with configuration flexibility when designing an LED driver for specific requirements. Hysteretic control is well-suited for applications such as light bulbs and traffic lights, in

which variable switching frequencies are tolerated or where narrow input-voltage range supplies are used. Hysteretic control doest not experience control-loop bandwidth restrictions, which eliminates the need for loop compensation because of its inherent stability. Utilizing hysteretic control to drive a buck-based LED driver (Fig. 2a) greatly simplifies the design as well as reduces the component count, and the cost of the driver. This configuration also yields superior PWM dimming ranges that outperform other buck-based schemes. Using hysteretic buck-based LED drivers with the shunt-dimming approach is well-suited for applications that require ultra-wide dimming ranges at high dimming frequencies and that can tolerate variable switching frequencies.

Fig. 2a: Basic hysteretic buck-based driver Quasi-hysteretic buck-based LED drivers offer a good compromise between fixed-frequency operation and hysteretic control for applications in which variable switching frequencies may not be desired. The controlled on-time (quasihysteretic) buck-based LED driver (Fig. 2b) employs a control scheme based on a hysteretic comparator and a oneshot on-timer which is used to set a controlled on-time. This controlled on-time is programmed so that it is inversely proportional to the input voltage, and, therefore, it minimizes the switching frequency variations as the line voltage changes. Using this scheme also eliminates the need for control-loop bandwidth limitations, enabling it to achieve wide dimming ranges when used with different dimming configurations.

Fig. 2b: Basic controlled on-time buck-based LED driver

In some cases, as in a number of automotive applications, synchronizing the LED driver(s) to an external clock or to each other may be required to minimize noise interference. Implementing the frequency synchronization feature with

the non-clock-based hysteretic and quasi-hysteretic scheme can be challenging. In contrast, this feature can be simply realized in clock-based regulators such as the fixed-frequency buck LED driver shown in Fig. 2c. Fixed frequency control generally yields a more complex solution, and it limits the dimming range of the driver regardless of the dimming approach due to its dynamic response limitations.

Fig. 2c: Basic fixed-frequency buck-based LED driver In summary, there are many characteristics that make buck-based regulators attractive LED drivers. They are simple to configure as a current source and can be realized with minimum component counts, which simplifies the design process, improves the drivers' reliability, and reduces cost. Buck-based LED drivers also provide configuration flexibility since they are compatible with multiple control schemes. They also allow for high-speed dimming as well as wide dimming ranges since they can be configured without output capacitance and are well-matched to various dimming approaches including shunt dimming. All these features make buck-based (step-down) LED drivers the topology of choice whenever the application permits. What if the application does not permit their use? Applications such as residential and commercial lighting require thousands of lumens, creating a need to drive LED strings. The total forward voltage drop of an LED string is equal to the sum of the forward voltage drops of all the LEDs in the string. In some cases, the input voltage range of the system can be lower than the forward voltage drop of the LED string, or it can vary so that sometimes it's lower and sometimes it's higher. These scenarios would require either boost, or buck-boost switching regulators. In the next installment, we discuss the challenges of using boost and buck-boost topologies to drive LEDs. About the authors Sameh Sarhan is a staff applications engineer for the Medium Voltage/High Voltage Power Management group in Santa Clara, CA. He has been involved with power electronics in various forms since 1998, having worked for FRC Corp. and Vicor Corp. His experience includes the design of hard/soft switching power supplies from a few watts to 600 watts. Sameh received a bachelor's degree in electronics engineering in 1996 from Cairo University (Egypt). Chris Richardson is an applications engineer in the Power Management Products group, Medium and High Voltage Division. His responsibilities are divided between lab work, bench evaluation of new ICs, written work such as datasheets and applications notes, and training for field engineers and seminars. Since joining National Semiconductor in 2001, Chris has worked mainly on synchronous buck controllers and regulators. In the last three years he has focused on products for the emerging high brightness LED market in the automotive and industrial areas. Chris holds a BSEE from the Virginia Polytechnic Institute and State University.

Design Article
A matter of light, Part 3---When to boost and buck-boost
Sameh Sarhan and Chris Richardson, National Semiconductor 5/31/2008 10:52 AM EDT

In part one of this series, we looked at the basics of LED lighting sources and their driving requirements. In part 2, we discussed why a constant-current buck converter should be your first preference when it comes to switch-mode LED drivers. In this third installment, we investigate larger LED displays and the applications space for other converter topologies. Manufacturers and designers of LED lighting often refer to applications with clear advantages for solid state illumination as "low-hanging fruit." Examples such as garden path lighting or MR16 bulb replacement often require only a few LEDs, or just one (Fig. 1). The most common voltages for low-voltage lighting are 12 VDC, 24 VDC, and 12 VAC. These applications often use a buck regulator. Although the buck is preferred, as previously discussed, the boost regulator is finding more use as the number of LEDs increase for LED lighting applications. Not content to pick off flashlights or single bulb replacements anymore, designers are targeting large-scale general illumination, systems that require thousands of lumens. Examples include street lighting, residential and commercial lighting, stadium lighting, and decorative or architectural lighting of spaces both interior and exterior. (Click on Image to Enlarge)

Figure 1: Buck and boost LED drivers with Vocalculation; buck: VO = n x VF, VO < VIN; boost: VO = n x VF, Vo > VIN Constant-current still required As with linear and buck-derived LED drivers, the main technical challenge in boost LED driver design is providing a controlled forward current, IF, to each LED of the array. Ideally every LED would be placed in a single series chain, ensuring that the same current flows through each device. A boost regulator is the simplest choice when stepping up a DC input voltage up to a higher DC output voltage, as it allows more LEDs to be placed in series from a given input voltage. A system designer for general illumination usually draws line power at 110 VAC or 220 VAC. If power factor correction (PFC), galvanic isolation, and line harmonic filtering aren't required, then single stage, non-isolated switching converters (buck, boost, or various buck-boost topologies) can use the rectified output of AC line voltage and directly drive long strings of series connected LEDs. In many cases, however, an intermediate DC bus voltage is used, derived from an AC/DC regulator that takes a universal AC input and provides PFC, isolation, and filtering. Besides solving legal requirements, a lower intermediate voltage bus reduces problems with dielectric breakdown, arcing, and improves the safety of service people working

with lighting. The European Union leads the world with the strictest legal requirements, including PFC for any lighting over 25 watts. North America and Asia consistently follow Europe's lead, albeit some years later. Safety standards and electrical codes such as UL and CE limit the output voltage of the AC/DC power supply that forms the input of the boost LED driver. Common rails are 12 and 24 volts, and in some cases 48 volts. Rarely are these intermediate bus rails higher than 60 volts, which is the cutoff for DC voltages under UL Class 2. The boost challenge Boost regulators are more difficult to design than buck regulators, regardless of whether we control the output voltage or the output current. The average inductor current in a continuous conduction mode (CCM) boost converter is equal to the load current (LED current) multiplied by 1/(1 - D), where D is the duty cycle. Boost voltage regulators require design review at the limits of input voltage to ensure correct design of the inductor, especially the peak current rating. A boost LED driver adds a variable output voltage that influences duty cycle and therefore the inductance and current rating of the main inductor. To prevent inductor saturation, the maximum average and peak currents must be evaluated at both VIN-MIN and VO-MAX. For example, over the range of process, drive current, and die temperature, a typical white InGaN LED's VF can vary from 3 to 4 volts. The more LEDs are placed in series, the greater the gap betweenVO-MIN and VO-MAX.

Unlike the buck regulator with its output inductor, the boost converter has a discontinuous output current. For this reason an output capacitor is required to keep the output voltage (and hence the output current) continuous. Where the output capacitor in a voltage regulator is designed to both filter and hold up the output voltage during load transients, in a current regulator it functions as an AC current filter only. The capacitance is made as low as possible, consistent with maintaining the desired LED ripple current. The lower the output capacitance (which keeps cost and size to a minimum), the faster the converter's response to changes in output current, and consequently the LED's dimming response is better. Another serious challenge for boost converters is the control loop. Buck regulators are available with voltage mode PWM control, peak current mode PWM control, constant/controlled on-time, and hysteretic control among others. Boost regulators in CCM (with the exception of low-power/portable equipment) are almost universally constrained to peak current mode PWM control, owing to their right-half plane zeroes and the fact that they deliver power to the output when the control switch is off. To design a boost LED driver that controls output current, the control loop must be analyzed using LEDs as the load, a case much different from the typical load of a boost voltage regulator. In peak current mode control, the impedance of the load has a strong effect on both the DC gain and the low-frequency pole of the control-to-output transfer function. For voltage regulators the load impedance is determined by dividing output voltage by output current. LEDs are diodes, with a dynamic resistance. This dynamic resistance can only be determined by plotting the VF versus IF curve and then taking the tangent line to find the slope at the desired forward current. As shown in Fig. 1, the current regulator uses the load itself as a feedback divider to close the control loop. This reduces the DC gain by a factor of (RSNS/ (RSNS + rD)). It is tempting to compensate a boost LED driver with a simple integrator, sacrificing bandwidth for stability. The reality is that many, if not most LED driver applications require dimming. Whether dimming is done by linear adjustment of IF (analog dimming) or by turning the output on

and off at high frequency (digital, or PWM dimming) the system requires high bandwidth and fast transient response just as a voltage regulator does. The buck-boost challenge LEDs for lighting are being adopted much faster than the standards for solid state illumination have developed. A wide variety of input voltages power a wide variety of LEDs. The number of LEDs in series, the type of LEDs, and the variation of VF with both process and die temperature all contribute to a wide range of output voltage. For example, high-end automobiles are converting to LEDs for their daytime running lamps. Three 3-watt white LEDs present a load of about 12 volts at a current of 1 amp. Automotive voltage systems usually require continuous operation over a range of 9 to 16 volts, with an extended range of 6 to 42 volts where performance is reduced but the system can operate without suffering damage. In general, the buck regulator makes the best LED driver, followed by the boost, but neither is appropriate for this case. If a buck-boost regulator must be used, the most difficult decision to make is often which topology to use. One fundamental difference between buck-boost regulators of any topology and the buck regulator or the boost regulator is that the buck-boosts never connect the input power supply directly to the output. Both the buck and the boost regulator connect VIN to VO (across the inductor and switch/diode) during a portion of their switching cycles, and this direct connection gives them better efficiency. All buck-boost regulators store the entire energy delivered to the load in either a magnetic field (inductor or transformer) or in an electric field (in a capacitor), which results in higher peak currents or higher voltage in the power switches. In particular, evaluation of the converter at the corners of both input voltage and output voltage is necessary because peak switch current occurs atVIN-MIN and VO-MAX, but peak switch voltage occurs at VIN-MAX and VIN-MAX andVO-MAX. In general this means that a buck-boost regulator of a certain output power will be larger and less efficient than a buck or boost regulator of equal output power. The single inductor buck-boost can be built with the same parts count as a buck regulator or boost regulator, making it attractive from a system cost standpoint. One disadvantage of this topology is that the polarity of Vo is inverted (Figure 2a) or regulated with respect to VIN (Figure 2b). Level-shifting or polarity inverting circuitry must be employed in these converters. Like the boost converter, they have a discontinuous output current, and require an output capacitor to maintain a continuous LED current. The power MOSFET suffers a peak current of IIN plus IF and a peak voltage of VIN plus VO. (Click on Image to Enlarge)

Figure 2: High-side buck-boost (a); low-side buck-boost (b) Other topologies The SEPIC converter has the advantages of a continuous input current due to the input inductor and positive output voltage. Like the boost and single inductor buck-boost it requires an output capacitor to maintain a smooth LED

current. A further advantage of SEPIC converters is that almost any low-side regulator or controller can be configured as a SEPIC without the need of polarity inversion or level shift circuitry.

Figure 3: SEPIC LED driver Rarely used in voltage regulation, the Cuk converter has emerged as an LED driver. Input and output currents are continuous. The polarity of the output voltage is reversed, as with the high-side buck-boost, but the output capacitor can be eliminated like the buck converter. The Cuk is the only other practical non-isolated regulator with this ability.

Figure 4: Cuk regulator Neither the boost nor the buck-boost regulator is preferred for switching LED drivers, owing to their higher complexity and parts count, lower efficiency (especially for the buck-boosts) and scant choice of control topologies. However, both are 'necessary evils' as LEDs push into more and more lighting applications. In some cases the system architecture can be altered to allow buck or even linear regulator-based LED drivers. Examples include very large light sources, such as street lights, where a hundred or more 1W+ LEDs are required. In general, LEDs for general illumination are working their way from lower to higher power, and in the intermediate arena, such as automotive headlights and small lighting fixtures, boost and buck-boost regulators represent the best choice for constant-current driving.

About the authors Sameh Sarhan is a staff applications engineer for the Medium Voltage/High Voltage Power Management group in

Santa Clara, CA. He has been involved with power electronics in various forms since 1998, having worked for FRC Corp. and Vicor Corp. His experience includes the design of hard/soft switching power supplies from a few watts to 600 watts. Sameh received a bachelor's degree in electronics engineering in 1996 from Cairo University (Egypt). Chris Richardson is an applications engineer in the Power Management Products group, Medium and High Voltage Division. His responsibilities are divided between lab work, bench evaluation of new ICs, written work such as datasheets and applications notes, and training for field engineers and seminars. Since joining National Semiconductor in 2001, Chris has worked mainly on synchronous buck controllers and regulators. In the last three years he has focused on products for the emerging high brightness LED market in the automotive and industrial areas. Chris holds a BSEE from the Virginia Polytechnic Institute and State University.

Design Article
A matter of light, Part 4 --- PWM dimming
Sameh Sarhan and Chris Richardson, National Semiconductor 6/5/2008 11:17 PM EDT

In part one of this series, we looked at the basics of LED lighting sources and their driving requirements. In part two, we discussed why a constant-current buck converter should be your first preference when it comes to switch-mode LED drivers. In part 3, we investigated larger LED displays and the applications space for other converter topologies. Here in the concluding part of this series, the authors take a look at how to best implement the dimming function. Whether you drive LEDs with a buck, boost, buck-boost or linear regulator, the common thread is drive circuitry to control the light output. A few applications are as simple as ON and OFF, but the greater number of applications call for dimming the output between zero and 100 percent, often with fine resolution. The designer has two main choices: adjust the LED current linearly (analog dimming), or use switching circuitry that works at a frequency high enough for the eye to average the light output (digital dimming). Using pulse-width modulation (PWM) to set the period and duty cycle (Fig. 1) is perhaps the easiest way to accomplish digital dimming, and a buck regulator topology will often provide the best performance. (Click on Image to Enlarge)

Figure 1: LED driver using PWM dimming, with waveforms PWM dimming preferred Analog dimming is often simpler to implement. We vary the output of the LED driver in proportion to a control voltage. Analog dimming introduces no new frequencies as potential sources of EMC/EMI. However, PWM dimming is used in most designs, owing to a fundamental property of LEDs: the character of the light emitted shifts in proportion to the average drive current. For monochromatic LEDs, the dominant wavelength changes. For white LEDs, the correlated color temperature (CCT) changes. It's difficult for the human eye to detect a change of a few nanometers in a red, green, or blue LED, especially when the light intensity is also changing. A change in color temperature of white light, however, is easily detected.

Most white LEDs consist of a die that emits photons in the blue spectrum, which strike a phosphor coating that in turn emits photons over a broad range of visible light. At low currents the phosphor dominates and the light tends to be more yellow. At high currents the blue emission of the LED dominates, giving the light a blue cast, leading to a higher CCT. In applications with more than one white LED, a difference in CCT between two adjacent LEDs can be both obvious and unpleasant. That concept extends to light sources that blend light from multiple monochromatic LEDs. When we have more than one light source, any difference between them jars the senses. LED manufacturers specify a certain drive current in the electrical characteristics tables of their products, and they guarantee the dominant wavelength or CCT only at those specified currents. Dimming with PWM ensures that the LEDs emit the color that the lighting designer needs, regardless of the intensity. Such precise control is particularly important in RGB applications where we blend light of different colors to produce white. From the driver IC perspective, analog dimming presents a serious challenge to the output current accuracy. Almost every LED driver uses a resistor of some type in series with the output to sense current. The current-sense voltage, VSNS, is selected as a compromise to maintain low power dissipation while keeping a high signal-to-noise ratio (SNR). Tolerances, offsets, and delays in the driver introduce an error that remains relatively fixed. To reduce output current in a closed-loop system, VSNS, must be reduced. That in turn reduces the output current accuracy and ultimately the output current cannot be specified, controlled, or guaranteed. In general, dimming with PWM allows more accurate, linear control over the light output down to much lower levels than analog dimming. Dimming frequency vs. contrast ratio The LED driver's finite response time to a PWM dimming signal creates design issues. There are three main types of delay (Fig. 2). The longer these delays, the lower the achievable contrast ratio (a measure of control over lighting intensity). (Click on Image to Enlarge)

Figure 2: Dimming delays As shown, tn represents the propagation delay from the time logic signal VDIMgoes high to the time that the LED driver begins to increase the output current. In addition, tsu is the time needed for the output current to slew from zero to the target level, and tsn is the time needed for the output current to slew from the target level back down to zero. In general, the lower the dimming frequency, fDIM, the higher contrast ratio, as these fixed delays consume a smaller portion of the dimming period, TDIM.The lower limit for fDIM is approximately 120 Hz, below which the eye no longer

blends the pulses into a perceived continuous light. The upper limit is determined by the minimum contrast ratio that is required. Contrast ratio is typically expressed as the inverse of the minimum on-time, i.e., CR = 1 / tON-MIN : 1 where tON-MIN = tD + tSU. Applications in machine vision and industrial inspection often require much higher PWM dimming frequencies because the high-speed cameras and sensors used respond much more quickly than the human eye. In such applications the goal of rapid turn-on and turn-off of the LED light source is not to reduce the average light output, but to synchronize the light output with the sensor or camera capture times. Dimming with a switching regulator Switching regulator-based LED drivers require special consideration in order to be shut off and turned on at hundreds or thousands of times per second. Regulators designed for standard power supplies often have an enable pin or shutdown pin to which a logic-level PWM signal can be applied, but the associated delay, tD, is often quite long. This is because the silicon design emphasizes low shutdown current over response time. Dedicated switching regulations for driving LEDs will do the opposite, keeping their internal control circuits active while the enable pin is logic low to minimize tD, while suffering a higher operating current while the LEDs are off. Optimizing light control with PWM requires minimum slew-up and slew-down delays not only for best contrast ratio, but to minimize the time that the LED spends between zero and the target level (where the dominant wavelength and CCT are not guaranteed). A standard switching regulator will have a soft-start and often a soft-shutdown, but dedicated LED drivers do everything within their control to reduce these slew rates. Reducing tSU and tSN involves both the silicon design and the topology of switching regulator that is used. Buck regulators are superior to all other switching topologies with respect to fast slew rates for two distinct reasons. First, the buck regulator is the only switching converter that delivers power to the output while the control switch is on. This makes the control loops of buck regulators with voltage-mode or current-mode PWM (not to be confused with the dimming via PWM) faster than the boost regulator or the various buck-boost topologies. Power delivery during the control switch's on-time also adapts easily to hysteretic control, which is even faster than the best voltage-mode or current-mode control loops. Second, the buck regulator's inductor is connected to the output during the entire switching cycle. This ensures a continuous output current and means that the output capacitor can be eliminated. Without an output capacitor the buck regulator becomes a true, high impedance current source, capable of slewing the output voltage very quickly. Cuk and zeta converters can claim continuous output inductors, but fall behind when their slower control loops (and lower efficiency) are factored in. Faster than the enable pin Even a pure hysteretic buck regulator without an output capacitor will not be capable of meeting the requirements of some PWM dimming systems. These applications need high PWM dimming frequency and high contrast ratio, which in turn requires fast slew rates and short delay times. Along with machine vision and industrial inspection, examples of systems that need high performance include backlighting of LCD panels and video projection. In some cases the PWM dimming frequency must be pushed to beyond the audio band, to 25 kHz or more. With the total dimming period reduced to a matter of microseconds, total rise and fall times for the LED current, including propagation

delays, must be reduced to the nanosecond range.

Consider a fast buck regulator with no output capacitor. The delays in turning the output current on and off come from the IC's propagation delay and the physical properties of the output inductor. For truly high speed PWM dimming, both must be bypassed. The best way to accomplish this is by using a power switch in parallel with the LED chain (Fig. 3). To turn the LEDs off, the drive current is shunted through the switch, which is typically an n-MOSFET. The IC continues to operate and the inductor current continues to flow. The main disadvantage of this method is that power is wasted while the LEDs are off, even through the output voltage drops to equal the current sense voltage during this time. (Click on Image to Enlarge)

Figure 3: Shunt FET circuit, with waveforms Dimming with a shunt FET causes rapid shifts in the output voltage, to which the IC's control loop must respond in an attempt to keep the output current constant. As with logic-pin dimming, the faster the control loop, the better the response, and buck regulators with hysteretic control provide the best response. Fast PWM with boost and buck-boost Neither the boost regulator nor any of the buck-boost topologies are well suited to PWM dimming. That's because in the continuous conduction mode (CCM), each one exhibits a right-half plane zero, which makes it difficult to achieve the high control loop bandwidth needed in clocked regulators. The time-domain effects of the right-half plane zero also make it much more difficult to use hysteretic control for boost or buck-boost circuits. In addition, the boost regulator cannot tolerate an output voltage that falls below the input voltage. Such a condition causes a short circuit at the input, and makes dimming with a parallel FET impossible. Among the buck-boost topologies, parallel FET dimming is still impossible or at best impractical due to the requirement for an output capacitor (the SEPIC, buckboost and flyback), or the uncontrolled input inductor current during output short circuits (Cuk and zeta). When true fast PWM dimming is required, the best solution is a two-stage system that uses a buck regulator as the second, LED driving stage. When space and cost do not permit this approach, the next best choice is a series switch (Fig. 4). (Click on Image to Enlarge)

Figure 4: Boost regulator with series DIM switch LED current can be shut off immediately. On the other hand, special consideration must be given to the system response. Such an open circuit is in effect a fast, extreme unloading transient that also disconnects the feedback loop and will cause the regulator's output voltage to rise without bound. Clamping circuits for the output and/or the error amplifier are required to prevent failure due to over-voltage. These clamps are difficult to realize with external circuitry, hence series FET dimming is practical only with dedicated boost/buck-boost LED driver ICs. In summary, proper control of LED lighting requires careful attention right from the start of the design process. The more sophisticated the light source, the more likely that PWM dimming will be used. This in turn requires the system designer to carefully consider the LED driver topology. Buck regulators offer many advantages for PWM dimming. If the dimming frequency must be high, or the slew rates must be fast, or both, then the buck regulator is the way to go. About the authors Sameh Sarhan is a staff applications engineer for the Medium Voltage/High Voltage Power Management group in Santa Clara, CA. He has been involved with power electronics in various forms since 1998, having worked for FRC Corp. and Vicor Corp. His experience includes the design of hard/soft switching power supplies from a few watts to 600 watts. Sameh received a bachelor's degree in electronics engineering in 1996 from Cairo University (Egypt). Chris Richardson is an applications engineer in the Power Management Products group, Medium and High Voltage Division. His responsibilities are divided between lab work, bench evaluation of new ICs, written work such as datasheets and applications notes, and training for field engineers and seminars. Since joining National Semiconductor in 2001, Chris has worked mainly on synchronous buck controllers and regulators. In the last three years he has focused on products for the emerging high brightness LED market in the automotive and industrial areas. Chris holds a BSEE from the Virginia Polytechnic Institute and State University.