The emergence of the emerging field of semiconductor lighting has made engineers who specialize in power electronics, optics and thermal management at the same time in demand.” title=”thermal management”>thermal management (mechanical engineering). There are not many engineers experienced in each field, and this usually means that the background of the system engineer or overall product engineer is related to these three fields, and they also need to cooperate with engineers in other fields as much as possible. The habits or accumulated experience that I have developed in the original field are brought into the design work, which is similar to what an electronic engineer who mainly studies digital systems encounters when he turns to solve the problem of power supply” title=”power supply” > power management: They may rely on pure simulation and route the board directly without testing the power supply on the bench because they fail to realize that switching regulators require careful inspection of the board layout; The working conditions are difficult to match with the simulation.

In the process of designing an LED luminaire, when the system architect is an Electronic power specialist, or when the power supply design is contracted to an engineering firm, some habits that are common in standard power supply design appear in LED driver design. A few habits are useful because LED drivers are very similar to traditional constant voltage sources in many ways. Both types of circuits operate over a wide input voltage range and high output power. In addition, both types of circuits face the challenges of connecting to an AC power source, a DC regulated power rail, or a battery. challenge.

Power electronics engineers are accustomed to always wanting to ensure high accuracy of output voltage or current, but this is not a good habit for LED driver design. Digital loads such as FPGAs and DSPs require lower core voltages, which in turn require tighter controls to prevent higher bit error rates. Therefore, the tolerances of digital power rails are usually controlled to within ±1% or less than their nominal values, and can also be expressed in absolute values, such as 0.99V to 1.01V. When the design habits of traditional power supplies are introduced into the field of LED driver design, the usual problem is: to achieve tight control of output current tolerance, more power will be wasted and more expensive components will be used, or both. Of.

cost pressure

The ideal power supply is inexpensive, 100% efficient, and takes up no space. Power electronics engineers are accustomed to hearing from customers, and they do their best to meet those requirements, trying to design systems within the smallest space and budget. It is no exception when it comes to LED driver design, in fact it faces greater budgetary pressures as traditional lighting technologies are fully commoditized and their prices are already very low. So, spending every penny of the budget is very important, and this is where some power electronics designers and engineers are “led astray” by old habits.

Controlling the LED current to the same accuracy as the supply voltage of the digital load would be a waste of electricity and cost. 100mA to 1A is the current range for most products today, especially 350mA (or rather, the current density of 350mA/mm2 for opto-semiconductor junctions) is a commonly adopted compromise between thermal management and lighting efficiency. The integrated circuits that control the LED drivers are silicon-based, so there is a typical bandgap in the 1.25 V range. To achieve 1% tolerance at 1.25V, a voltage range of ±12.5mV is required. This is not difficult to achieve, and there are many inexpensive voltage reference circuits or power control ICs that can achieve this tolerance or better. When controlling the output voltage, a high precision resistor can be used to feedback the output voltage at very low power (as shown in Figure 1a). To control the output current, some adjustments to the feedback method are required, as shown in Figure 1b. This is currently the only and easiest means of controlling the output current.

Figure 1a: Voltage feedback; Figure 1b: Current feedback

A major disadvantage of this approach is that both the load and feedback circuits are identical. The reference voltage is applied across a resistor in series with the LED, which means that the higher the reference voltage or LED current, the more power the resistor will dissipate. Therefore, the reference voltage of the first generation of dedicated LED driver ICs is much lower than that of current products, which is similar to a battery charger. Lower voltage means lower power dissipation, which also means smaller, cheaper, lower loss current sense resistors. In the simple low-side feedback environment shown in Figure 1b, 200mV is a conventional voltage choice. However, achieving a tolerance of ±1% at a 200mV reference requires a very expensive integrated circuit, in which case the tolerance relative to the nominal reference is ±2mV. While this is not impossible, higher accuracy comes at a higher cost. A tolerance of ±2mV requires the production, testing, and binning techniques required for high-accuracy voltage references, at which point the additional cost should be spent on smarter LED drivers. The value of the new expense is the addition of a feedback loop whereby the light output (instead of the current output) can be used to control how the LED is driven.

Measuring light output

Just as digital product designers use simulation to solve uncertain problems in power supply design, system architects who are power electronics engineers think of high-precision output when designing LED lamps. LED manufacturers have clearly shown that the luminous flux is proportional to the forward current. Drive all LEDs with the same current, so each LED will produce the same luminous flux. Therefore, power electronics engineers come to the conclusion that high-precision current flow is necessary. In doing so, they forget that the lumens and lux values ​​(not amps) of the light output are the point. Measuring current is easy, while measuring light requires expensive and large equipment, such as the integrating sphere shown in Figure 2, which most electronics engineers don’t know much about.

Figure 2: Sectional view of the optical integrating sphere

Also, even though a ±0.1% tolerance current source (which can be quite expensive) has a huge market value, it does little to produce a tight tolerance value in the actual light output. This can be determined by looking at the binning of the LED luminous flux. Table 1 gives the world’s three top power optoelectronic semiconductor manufacturers’ high-end cool white LEDs at 350mA and 25? Luminous flux binning results under C condition. Note that the last column is the tolerance average for each bin, not the tolerance across all luminous flux bins.

Table 1 High-end cool white LEDs of the world’s three top power optoelectronic semiconductor manufacturers at 350 mA and 25? Luminous flux binning results under C.

The emergence of the emerging field of semiconductor lighting has made engineers who specialize in the three fields of power electronics, optics and thermal management (mechanical engineering) in demand. Currently, there are not many engineers experienced in all three fields, and this usually means that the background of system engineers or overall product engineers is related to these three fields, and they need to cooperate with engineers in other fields as much as possible. Oftentimes, systems engineers bring their own habits or accumulated experience into their design work, similar to what happens when an electronics engineer who studies primarily digital systems turns to solving power management problems: they may rely on Pure simulation, without testing the power supply on the test bench and routing directly on the circuit board, because they do not realize: switching regulators need to carefully check the circuit board layout; In addition, if not tested on the test bench, the actual operation It’s hard to be consistent with simulation.

In the process of designing an LED luminaire, when the system architect is an electronic power specialist, or when the power supply design is contracted to an engineering firm, some habits that are common in standard power supply design appear in LED driver design. A few habits are useful because LED drivers are very similar to traditional constant voltage sources in many ways. Both types of circuits operate over a wide input voltage range and high output power. In addition, both types of circuits face the challenges of connecting to an AC power source, a DC regulated power rail, or a battery. challenge.

Power electronics engineers are accustomed to always wanting to ensure high accuracy of output voltage or current, but this is not a good habit for LED driver design. Digital loads such as FPGAs and DSPs require lower core voltages, which in turn require tighter controls to prevent higher bit error rates. Therefore, the tolerances of digital power rails are usually controlled to within ±1% or less than their nominal values, and can also be expressed in absolute values, such as 0.99V to 1.01V. When the design habits of traditional power supplies are introduced into the field of LED driver design, the usual problem is: to achieve tight control of output current tolerance, more power will be wasted and more expensive components will be used, or both. Of.

cost pressure

The ideal power supply is inexpensive, 100% efficient, and takes up no space. Power electronics engineers are accustomed to hearing from customers, and they do their best to meet those requirements, trying to design systems within the smallest space and budget. It is no exception when it comes to LED driver design, in fact it faces greater budgetary pressures as traditional lighting technologies are fully commoditized and their prices are already very low. So, spending every penny of the budget is very important, and this is where some power electronics designers and engineers are “led astray” by old habits.

Controlling the LED current to the same accuracy as the supply voltage of the digital load would be a waste of electricity and cost. 100mA to 1A is the current range for most products today, especially 350mA (or rather, the current density of 350mA/mm2 for opto-semiconductor junctions) is a commonly adopted compromise between thermal management and lighting efficiency. The integrated circuits that control the LED drivers are silicon-based, so there is a typical bandgap in the 1.25 V range. To achieve 1% tolerance at 1.25V, a voltage range of ±12.5mV is required. This is not difficult to achieve, and there are many inexpensive voltage reference circuits or power control ICs that can achieve this tolerance or better. When controlling the output voltage, a high precision resistor can be used to feedback the output voltage at very low power (as shown in Figure 1a). To control the output current, some adjustments to the feedback method are required, as shown in Figure 1b. This is currently the only and easiest means of controlling the output current.

Figure 1a: Voltage feedback; Figure 1b: Current feedback

A major disadvantage of this approach is that both the load and feedback circuits are identical. The reference voltage is applied across a resistor in series with the LED, which means that the higher the reference voltage or LED current, the more power the resistor will dissipate. Therefore, the reference voltage of the first generation of dedicated LED driver ICs is much lower than that of current products, which is similar to a battery charger. Lower voltage means lower power dissipation, which also means smaller, cheaper, lower loss current sense resistors. In the simple low-side feedback environment shown in Figure 1b, 200mV is a conventional voltage choice. However, achieving a tolerance of ±1% at a 200mV reference requires a very expensive integrated circuit, in which case the tolerance relative to the nominal reference is ±2mV. While this is not impossible, higher accuracy comes at a higher cost. A tolerance of ±2mV requires the production, testing, and binning techniques required for high-accuracy voltage references, at which point the additional cost should be spent on smarter LED drivers. The value of the new expense is the addition of a feedback loop whereby the light output (instead of the current output) can be used to control how the LED is driven.

Measuring light output

Just as digital product designers use simulation to solve uncertain problems in power supply design, system architects who are power electronics engineers think of high-precision output when designing LED lamps. LED manufacturers have clearly shown that the luminous flux is proportional to the forward current. Drive all LEDs with the same current, so each LED will produce the same luminous flux. Therefore, power electronics engineers come to the conclusion that high-precision current flow is necessary. In doing so, they forget that the lumens and lux values ​​(not amps) of the light output are the point. Measuring current is easy, while measuring light requires expensive and large equipment, such as the integrating sphere shown in Figure 2, which most electronics engineers don’t know much about.

Figure 2: Sectional view of the optical integrating sphere

Also, even though a ±0.1% tolerance current source (which can be quite expensive) has a huge market value, it does little to produce a tight tolerance value in the actual light output. This can be determined by looking at the binning of the LED luminous flux. Table 1 gives the world’s three top power optoelectronic semiconductor manufacturers’ high-end cool white LEDs at 350mA and 25? Luminous flux binning results under C condition. Note that the last column is the tolerance average for each bin, not the tolerance across all luminous flux bins.

Table 1 High-end cool white LEDs of the world’s three top power optoelectronic semiconductor manufacturers at 350 mA and 25? Luminous flux binning results under C.

Calculated light output accuracy

Knowing that the LED light output from a single flux bin will have a tolerance of ±3% to ±10%, a system engineer may conclude that the drive current tolerance value must be as tight as possible. However, from a statistical point of view, this view is not correct. A common but incorrect assumption is that the overall tolerance of any value is equal to a simple accumulation of the values ​​under worst-case conditions. The tolerance of the current source powering the LED and the tolerance of the LED luminous flux are independent of each other – they are independent of each other from the very beginning. For two uncorrelated factors X and Y, the overall tolerance Z is not the sum of the tolerances of X and Y, but should be calculated using the following expression:

Table 2 and Figure 3 provide a comparison of the overall tolerance and the assumed current source tolerance, assuming that the LED light output varies linearly with forward current in the 350mA region.

Table 2 Comparison of overall tolerance and hypothetical current source tolerance.

Figure 3: Overall tolerance versus hypothetical current source tolerance.

According to equation (1), it can be found that the effect of the lowest tolerance factor is greater than that of the others, and the actual overall tolerance value is much better than the worst-case tolerance sum of each factor, especially when one factor is much better than the other factor time. As can be seen from Figure 3, the most reasonable goal for the current source tolerance is to keep it within the tolerance of the LED light output. Remember: Many fixtures use LEDs from different bins for cost reasons. Table 3 lists the tolerance values ​​under the highest two, three, and four luminous flux bins of the same LED.

Table 3 Tolerance values ​​of the highest two, three and four luminous flux levels of the same LED

Dimming control

LED manufacturers and their distribution partners are working hard to improve the luminous flux tolerance of their products, offering finer binning at a reasonable cost. For designers who want a product that lasts 5 years or 50,000 hours and maintains overall light output over the lifespan, even the densest flux binning and 0.1% tolerance Current sources are also difficult to implement. Because two important factors, heat and performance degradation over time, reduce the luminous flux of the LED, even a current source tolerance and LED luminous flux tolerance of 0.001% will not solve the problem. Given these losses, designers of high-quality solid-state lighting products must find power supplies with additional feedback loops, namely heat and light sources. For this purpose, dimming control is required, and those integrated circuits that can perform linear control and PWM control on the output current become the best choice.

For example, National Semiconductor’s LM3409 and LM3424 are LED driver control ICs, which are second-generation current sources for semiconductor lighting. Both products can control the average LED current value through a variable resistor or voltage source, and provide a dedicated input signal for the PMW dimming signal. In addition to the linear control loop, the analog regulation capabilities of the LM3409 and LM3424 also allow system designers to make their own choices between the trade-offs between output current accuracy and size, cost, and current sense resistor power dissipation. The LM3409/09HV control buck circuit shown in Figure 4 is the most commonly used circuit mode in power LED drivers. The LM3424 in Figure 5 can be used as a boost regulator LED driver, as well as a buck/boost, SEPIC, flyback, or even a “floating” buck circuit.

Figure 4: LM3409/09HV Buck LED Driver

Figure 5: LM3424 Boost LED Driver

Applications requiring light control

A street light is a good example of a light source because it has strict legal standards. For road lights, EU countries stipulate their minimum and maximum light output and lighting modes. For LED street lights that comply with this regulation and provide a lifespan of five years or more, the design must take into account the immediate luminous flux loss due to heat, as well as the flux loss due to performance degradation over longer periods of time.

A natural approach is to use a light sensor, such as a photodiode that forms a linear control loop. On day one of the system, only a fraction of the overall available drive current should be used, taking into account that over time the drive current will slowly increase to an upper limit, thereby keeping the light output constant. The photodiode can be biased and converted into a pulse width modulated signal, which will help maintain a more constant relative color temperature in the dimming range, its linear control loop is simpler, and the dimming range is generally smaller. PWM control will be more useful when controlling light output based on different timing, motion sensors or other power saving measures. Figure 6 presents a hypothetical block diagram of an LED lamp with longer lifetime and constant light output.

Figure 6: PWM (Pulse Width Modulation) is used for day/night control and linear control is used for light output.

Summary of this article

Output current accuracy is only one aspect of LED driver performance, but when the luminous flux tolerance of the LED itself remains well above ±1%, even if the current source tolerance is the same as the voltage rail tolerance in the digital processor Strict, and almost meaningless, the average LED current tolerance should be nearly equal to the luminous flux tolerance. This article explores an ideal situation based on the error of a single bin, and gives some practical examples using two or more bins of LEDs that can also more easily achieve tolerances of ±5%, ± 10% or higher. In the additional control loop, the cost overhead should be used for 1% current control, and power can be used for higher sense voltages. Some LED lights will put more emphasis on simplicity, practicality and low cost. At this time, even linear dimming will be too complicated and expensive. However, if you want to design a lamp that can play the full performance of LED, you need to use linear control or PWM method or both. Coordinated use to optimize product performance and longevity.

Christopher Richardson

system application engineer

National Semiconductor