The Subtle Circuitry Behind LED Lighting

The circuitry behind LED lighting poses tricky challenges

9 min read
Illustration of a light bulb.
Illustration: David Arky

Last August, the U.S. Department of Energy announced the first winner in its ongoing competition to encourage lighting that’s more efficient—the Bright Tomorrow Lighting Prize, or L Prize. The DOE awarded Philips Lighting North America US $10 million for coming up with a lamp that’s equivalent to a standard 60-watt incandescent bulb in size and brightness but lasts at least 25 times as long and runs on less than 10 W.

Although lamps that are almost as efficient have been available for more than a year, the prizewinning design is just now going on sale. Like the backlights in modern cellphones and computer monitors, these lamps use light-emitting diodes to generate white light. They offer long lifetimes, pleasing colors, and most important, phenomenal energy efficiency.

Is it now time to throw away the incandescent bulbs still lurking in your light fixtures—and even the compact fluorescent lamps (CFLs) you’ve been switching to—and replace them all with LED superlights? With costs often hovering around $25 a pop, few home­owners are rushing to take that plunge. But prices are dropping, and performance is improving fast. So it’s clear that the day when LED lamps will dominate lighting in both residences and businesses is not far off.

Why are LED-based lamps superior, and what makes them so tricky to engineer, anyway? You might imagine that the answers would hinge on the subtleties of solid-state semiconductor physics that govern high-brightness LEDs. They do, but only up to a point. The practicality of these new lights also depends on a more mundane part of the package that’s often overlooked: the circuitry required to drive them. Here I’ll explain what the requirements are for that circuitry and why designing the appropriate electronics can be a challenge, although not one that should slow the adoption of this fantastic new form of lighting.

Like it or not, incandescent bulbs are a dying breed. Australia and the European Union started phasing out traditional incandescents in 2009. The United States is haltingly moving in the same direction, and China is aiming to eliminate incandescent bulbs by 2016. The reason is simple: Old-fashioned lightbulbs squander enormous amounts of electricity.

A full 90 percent of the energy you put into an ordinary incandescent bulb goes into making heat, not light. A standard 60-W bulb generates approximately 850 lumens of light, which comes out to about 14 lumens per watt. Halogen lamps (a more sophisticated kind of incandescent with a higher temperature filament) can provide about 20 lm/W. CFLs are considerably more efficient, producing around 60 lm/W, but they have other problems.

One common complaint is that you can’t dim them. (In truth, some can be dimmed, but their range is usually limited.) Also, CFLs are slow to light up, and because their bulbs contain mercury vapor, they present an environmental hazard. Even with recycling opportunities available, millions of these bulbs end up in landfills every year.

LED-based lights have none of those drawbacks, and they are far more efficient, some offering more than 100 lm/W. These nominally white lights, in fact, contain blue LEDs, along with a phosphor coating that converts the narrow wavelength light they emit into something the human eye perceives as white. With the appropriate mix of phosphor materials, designers can set the tone of the light from cool to warm, depending on the application they have in mind.

Next to their high energy efficiency, the most attractive quality of LED lights is their longevity. Exactly how long one will last depends on how it’s designed and operated, but most will work for 25 000 hours or more while maintaining at least 70 percent of their initial light output. And many manufacturers advertise 35 000-hour lifetimes. So if you used an LED lamp for 10 hours a day, you could expect it to last from 7 to almost 10 years. That’s a far cry from a standard incandescent bulb, which on average goes dark after only about 1000 hours of use. It also beats CFLs, which typically last from 6000 to 10 000 hours.

Such long lifetimes reduce one of the hidden costs of lighting, especially for commercial and industrial users: maintenance and replacement costs. That, and the energy savings that accrue, explains why large-scale users have been the early adopters. For example, the city of Los Angeles is now in the process of replacing 140 000 high-pressure sodium streetlights with LEDs. Major retailers, like Walmart and McDonald’s, are also switching to LED lighting in some places. Really, the only thing holding such businesses back is the high up-front costs—and the prospect that LED lighting technology will soon improve and become an even better deal.

One drawback of the LED, however, is that, unlike an incandescent bulb, it can’t just run straight off the electric mains. The operating voltage of a standard white-light LED is usually in the range of 3 to 3.6 volts, about the same voltage as the lithium-ion battery in your cellphone. Although this makes LEDs easy to use in mobile devices, most lighting fixtures get power from the grid. So conversion circuitry is required to transform the AC line voltage into a form that can drive individual LEDs.

The necessary circuitry is similar to that in a cellphone charger or laptop adapter, with some key differences. First, because LEDs can operate for many years, the power electronics that drive them must either last just as long or be configured so that any failure-­prone circuits can easily be replaced. Also, because the drive electronics must often be embedded within a screw-in light source, the circuitry must be very compact. It should also be energy efficient, because any losses from the drive electronics increase the total power that must be drawn from the wall outlet. Lastly, and rather surprisingly, the drive circuitry must be able to withstand relatively high operating temperatures. 

That last statement requires some explanation. As I’ve noted, incandescent bulbs turn only 10 percent of the electrical energy they consume into light, with the rest wasted as heat. LEDs transform about 50 percent of the energy fed to them into light, making them far more efficient. But there’s a ­complication: Incandescent bulbs radiate their waste heat into the space around them as infrared waves, whereas LEDs radiate only visible light. Also, the ceramic bases of screw-in LED lamps act as insulators. So their waste heat, modest as it is, tends to remain at the source. That spells trouble, for a couple of reasons.

For one, the heating causes the temperature of the LEDs to rise—and here, hotter isn’t better. Light output drops as the temperature of the lamp increases (just the opposite of what happens with fluorescent lights). Worse, high tempera­tures shorten the life of LEDs. Another problem is that as the drive ­circuitry heats up, various electronic components—particularly electrolytic ­capacitors—wear out faster.

One way system designers combat those problems is to use a metal heat sink, allowing convection to shed heat into the environment. Another is to avoid ­creating any more waste heat than is absolutely necessary by designing drive circuitry that’s highly efficient.

Although specialized circuits are sometimes attached to individual LEDs, more often than not one set of drive electronics powers multiple LEDs wired together. Indeed, some LED manufacturers mount an array of LEDs in an integrated package to achieve higher light output, although single high-­output LEDs are also common.

Image: Emily Cooper
Internal Affairs: LED lamps contain an assortment of highly engineered components. The generic example shown here includes an array of white-light LEDs and the electronic circuitry to drive them, all packaged in a compact screw-in unit.
Click on image for a larger view.

In most instances, the individual LEDs in each group are wired in series. Connecting them this way ensures that the same amount of current flows through each one, even if there are minor differences in their electrical characteristics. And that’s exactly what you want, because drive current determines their light output and color. So you need to do all you can to maintain the specified current level.

This need for a constant current does not exist in most electronic devices. A microprocessor, for example, accepts a fixed voltage and, depending on what task it is performing, draws more or less current. You can’t, however, just apply a fixed voltage to an LED and expect a set amount of current to pass through it. That’s because the voltage across the diode varies with temperature and also with the amount of current it draws. Also, there can be considerable manufacturing variation between LEDs, not to mention variation between similar devices from different suppliers.

Often, though, it’s not practical to wire all the LEDs you need into one big series-connected chain. For the desired amount of light, you might need so many LEDs that the voltage to drive them would become excessive if you had them all wired in series. The obvious solution is to limit the number of LEDs in each string and to power several strings in parallel if need be.

That’s straightforward if each string has its own drive circuit, but if several strings share the same power source, life gets more complicated. For one thing, putting LEDs in parallel requires that the components be well matched—­otherwise the current (and light output) in each string won’t be the same. And there’s the danger that one LED will fail and cut off the flow of electricity through the string it’s in, like what used to happen frustratingly often with old-­fashioned Christmas-tree bulbs. That’s bad, of course, because the whole string goes dark. Also, it will send more current into the parallel strings, which increases their temperature and will damage them if the current is too high. Designers can avoid such cascading failures, however, by cross-­connecting the LEDs in parallel strings. A single point of failure would then affect only a few other LEDs.

Ideally, though, each series-connected string would have its own regulated driver, providing just the right amount of current needed. LED manufacturers carefully document the amount of current required for a given light output, so it’s not hard to decide what current to provide. The voltage needed to maintain that current level might vary from, say, 3 to 3.6 volts. So if, for example, eight LEDs are wired in series within one lamp, the drive circuitry for it must provide the desired current level at voltages that range from 24 to about 29 volts.

The drive electronics must include two basic functional elements: a power conversion circuit (essentially a transistor switch that rapidly turns on and off) and a sensing circuit, which monitors the average current through the LEDs and provides a feedback signal to regulate the proportion of time that the power-conversion switch remains turned on. In many cases, a transformer is used to change voltages and to isolate the LED from the high-voltage electrical mains. In such designs, the feedback signal is often communicated optically from the sensing electronics to the power-conversion circuitry, so as not to compromise the electrical isolation between these two stages.

Arranging all this is easy enough for engineers versed in designing switched-mode power supplies, like those inside cellphone chargers or desktop computers. One looming challenge with LED lighting, though, is that it promises to make switched-mode power supplies even more widespread than they are now. This is great for companies like the one I work for, On Semiconductor, based in Phoenix, which builds ICs for use in such supplies. But it could present a headache for electric utilities, unless additional measures are taken to ensure that these power supplies are grid friendly. Let me explain.

The amount of current an ordinary incandescent lightbulb draws at any given instant is proportional to the voltage applied across it. As the magnitude of this AC voltage oscillates, so does the current flowing through the bulb, along with the energy expended. As a result, the power that the local utility company generates flows smoothly into the bulb where it’s converted to light and heat.

Many electrical loads, however, contain capacitors or inductors, which can store energy and thus alter how the device draws current from the electrical mains. Substantial capacitance or inductance will push the timing of the voltage and current oscillations out of whack, allowing energy to flow back and forth between the load and the grid. Another problem is the generation of harmonics of the grid’s fundamental frequency.

Power companies can deal with these disruptions, but they are nevertheless troublesome. This is why regulatory authorities are attempting to limit the problems LED lighting might create. The usual gauge for judging that is called the power factor, which varies from 0 (when energy just flows back and forth without being consumed) to 1 (when all the energy flows smoothly into the load). In the United States, for example, any LED bulb that draws more than 5 W, or any LED-based lighting fixture intended for residential use, must have a power factor greater than 0.7 to qualify for an Energy Star rating. And LED fixtures intended for commercial use must have power factors greater than 0.9 to qualify.

The adoption of LEDs for general lighting will no doubt be both evolutionary and revolutionary. On one hand, many people will shift to LEDs bit by bit, using the lamps they have always used and simply purchasing replacements for their screw‑in incandescents and CFLs. On the other hand, LEDs present designers with ways to create much more innovative forms of lighting, ones that take advantage of the long lifetime, directionality, and fine-grain scalability of light that LEDs offer. Lighting designers for homes and businesses will need time to discover the possibilities, but once they do, fantastic new kinds of lighting will surely start to illuminate our homes and offices. And if the circuits that operate them are built right, those lights will prove just as reliable as they are attractive.

This article originally appeared in print as “Driving the 21st Century’s Lights.”

About the Author

Bernie Weir, an application and marketing manager at On Semiconductor, earned his EE degree from the Rose-Hulman Institute of Technology. He began working with the electronics that drive LED lamps in the early 2000s, but only in the past few years have technical developments and industry standardization for LED lighting come together, he says.

About the Photographer

For more about the X-ray photos in this article, see the Back Story,  “Penetrating Insight.”

The Conversation (0)

Video Friday: DARPA Subterranean Challenge Final

1 min read

This week we have a special DARPA SubT edition of Video Friday, both because the SubT Final is happening this week and is amazing, and also because (if I'm being honest) the SubT Final is happening this week and is amazing and I've spent all week covering it mostly in a cave with zero access to Internet. Win-win, right? So today, videos to watch are DARPA's recaps of the preliminary competition days, plus (depending on when you're tuning in) a livestream of the prize round highlights, the awards ceremony, and the SubT Summit with roundtable discussions featuring both the Virtual and Systems track teams.

Keep Reading ↓ Show less

Making 3D-Printed Objects Feel

3D-printing technique lets objects sense forces applied onto them for new interactive applications

2 min read

Researchers from MIT have developed a method to integrate sensing capabilities into 3D printable structures comprised of repetitive cells, which enables designers to rapidly prototype interactive input devices.


Some varieties of 3D-printed objects can now “feel," using a new technique that builds sensors directly into their materials. This research could lead to novel interactive devices such as intelligent furniture, a new study finds.

The new technique 3D-prints objects made from metamaterials—substances made of grids of repeating cells. When force is applied to a flexible metamaterial, some of their cells may stretch or compress. Electrodes incorporated within these structures can detect the magnitude and direction of these changes in shape, as well as rotation and acceleration.

Keep Reading ↓ Show less

Safety in the Spotlight: Hendrix Covered Conductors

Safety-related contact currents for humans touching tedium voltage covered conductors

1 min read
Hendrix Aerial Cable Systems

This white paper explores what happens when humans come into physical contact with medium voltage covered conductors under a range of conditions. Safety thresholds, lab tests of charging currents, and CEATI testing of covered conductors are discussed. Authored by experts at Hendrix Aerial Cable Systems.

Trending Stories

The most-read stories on IEEE Spectrum right now