Bumps in the road ahead for solid-state lighting

Bumps in the road ahead for solid-state lighting

Trying to shoehorn a cutting-edge technology into an over-100 year-old infrastructure will not be easy or linear

BY GEORGE KELLY
Illumineer
Avnet Electronic Marketing
www.em.avnet.com

LED retrofit lamps have started to show up on retail shelves and many in the LED industry are giddy about the anticipated growth over the next few years. Nevertheless, the road to success for LED lighting will not be completely smooth, as there are a number of pitfalls to navigate.

Trying to shoehorn a cutting-edge technology into an infrastructure that was developed over 100 years ago will not be easy or linear. Several issues are solved or are about to be solved, but several challenges remain, many of which offer opportunities for those with the answers. Some of the potential problems ahead are pricing, color quality, thermal management, regulatory, and consumer education. While most of these issues are not technical, the design engineer will nonetheless need to understand them, if not face them head on. These issues will vary depending on whether you are designing a retrofit light bulb or a custom LED luminaire.

LED retrofit bulbs

In a few months, the 100-W tungsten incandescent light bulbs will be effectively banned by the federal government, and compact fluorescent bulbs (CFLs) are expected to be the initial beneficiary of this ban. But one has to wonder why consumers have not switched over to CFLs for purely their own financial interests due solely to the real energy savings of CFLs over incandescent bulbs. Some may chalk up the failure of CFLs to destroy incandescent bulbs in the marketplace to consumer ignorance, but shouldn’t the millions spent on marketing CFLs have taken care of that years ago?

The fact is that a significant percentage of consumers have rejected CFLs for myriad reasons, from flicker, color rendering, reliability versus cost, slow startup, and environmental concerns about the mercury they contain. While CFLs have lessened if not eliminated many of these issues, in the minds of many consumers CFLs are still undesirable. In the short term, many of these consumers will turn to halogen incandescent bulbs as their 100 W tungsten incandescent bulbs burn out.

This raises the question: can LEDs jump into the vacuum and become a viable option for consumers? Currently there are no commercially available LED retrofit bulbs to replace the ubiquitous 100-W A19 incandescent bulb. The reason for this is that it is currently impossible to passively cool an LED that outputs 1,500 lm (the typical output of a 100-W light bulb) in the physical confines of the A19 form factor.

While white LEDs are very efficient light sources, converting approximately one-third of the input power into light, the remaining two-thirds is converted into heat in the LED. So at 100 lm/W, about 10 W must be continuously and rapidly dissipated. If this heat is not removed, the LED will quickly heat up beyond its maximum operating temperature of approximately 120ºC.

This problem will likely be solved eventually by improvements in the luminous efficiency of LEDs. Last year, the best cool white LEDs achieved 100 lm/W luminous efficiency; this year, several cool white LEDs are available in the 120 to 130 lm/W range. This trend is expected to continue unabated for the foreseeable future.

As LED efficiencies improve, the thermal management improves by approximately the square of the efficiency, because the total power supplied to the LED decreases and the percentage of heat generated by that input power also decreases by the same amount.

The rapid improvement in LED luminous efficiency is both a boon and a bane. Savvy customers who were generally also the earlier adopters may want to wait for the technology to mature before buying especially given the current high price of LED bulbs. Today, 75-W PAR38 retrofit bulbs are priced in the $40 range for most consumers this is excessive, considering they can purchase a CFL for under $5 and are used to buying incandescents for pennies.

Yet, given the 10- to 15-year lifetime of LED bulbs, $40 dollars is a reasonable price. But the consumer must learn to see LEDs bulbs as a capital improvement to their home, rather than as a consumable as light bulb purchases have always been. Will most consumers make this shift in thinking when their home values are flat or still falling? Maybe not initially, but consumers interest in energy conserving technologies could grow as we see an influx of marketing campaigns that explain the many advantages of LEDs.

Avoiding instant obsolescence

LED bulb manufacturers also must face the prospect that their new light bulb may be obsolete before it reaches store shelves. With SSL rapidly improving, time-to-market is critical. While actual design times can be relatively short, obtaining the required certifications from UL to Energy Star can take months.

Design engineers should be careful to select LEDs that will be manufactured for years to come. LED manufacturers are good about releasing higher luminous flux bins as they are developed in the same package. Packaging technology is also advancing as quickly as LED die technology and can be as important to LED performance improvements as die technological development. LED manufacturers understand this issue, and in response are extending their product life cycles and, when possible, introducing new LEDs with the same PCB footprint as previous models.

Color quality may be the most difficult problem to solve. The industry has spent tremendous time and expense in measuring and controlling the color variability of white LEDs. Chromaticity binning has become more and more precise with manufacturers like OSRAM offering ANSI fine binning. But color temperature and tight chromaticity binning don’t tell the complete story, because two light sources with identical chromaticity coordinates may have very different wavelength spectra. If the spectra are too different, non-white surfaces will appear to be different colors under the two light sources.

This is generally not an issue for two LED lights with the same chromaticity coordinates, because their underlying spectra will be very similar. It is, though, a potential issue for an LED light replacing an incandescent light with the same color temperature (and consequently the same chromaticity coordinates), since the LED spectrum is very different from the incandescent’s spectrum.

The Color Rendering Index or CRI is a measure of how closely the perceived color of a surface illuminated by a particular light source will be to the perceived color of the same surface under incandescent illumination. A CRI of 100 is a perfect match. A CRI above 80 for an LED is considered good, but consumers that are more discriminating may beg to differ.

Fig. 1. By adding red LEDs to the usual greenish-white ones, OSRAM’s Brilliant Mix provides a closer match to the incandescent light spectrum.

OSRAM has introduced their Brilliant Mix technology that adds red LEDs to greenish-white LEDs (see Fig. 1 ). When the red and greenish-white light mix, the result is a white spectrum that better matches the incandescent spectrum especially at the red end of the spectrum where conventional white LEDs typically drop off. The result is a high luminous efficiency with a high CRI.

LED luminaires

In the short term, LED retrofit bulbs will make the initial splash, but in the long term there are great opportunities for custom LED luminaires, where the constraints imposed by the 100-year-old electrical infrastructure can be avoided.

LEDs offer tremendous advantages over both incandescents and CFLs beyond just raw lumens/W. LEDs make possible much more complex form factors and consequently can create more interesting and useable illumination patterns than traditional bulbs and CFLs. To take advantage of this, a custom luminaire is necessary. LEDs also offer additional features such as tunable color temperature, enhanced brightness and intelligent controls.

Imagine a luminaire that not only is dimmable, but one that you can select the color temperature you desire. Alternatively, you may want to enhance the saturation of one or more colors in the room. LEDs are the only light source whose spectral characteristics are flexible enough to make such a luminaire. While this is technically possible, creating a user interface for making these adjustments that is intuitive will be challenging. The manufacturer that can design a color-tunable luminaire with simple controls will have a unique competitive advantage.

Another possibility is to raise the illumination level, especially in office spaces. Our eyes start to lose their ability to focus after age 40; even those who had perfect eyesight throughout their life will require reading glasses at some point in their forties or fifties. In many cases, though, these people do not need reading glasses when reading outdoors, because our eyes see more clearly in bright illumination levels. When LED efficiencies reach the 150 lm/W range, it will become feasible to increase office lighting to 1,000 lux, as opposed to the 300 lux now typical in most office spaces.

More opportunities arise when one considers other constraints of conventional lighting. Typical home lighting requires several 20-A circuits per home. If all of the ceiling light fixtures are converted to LEDs, a single 10-A circuit (or several low amperage circuits adding up to 10 A) would be more than enough to light the whole house. And since LEDs are inherently low voltage devices the design engineer can entertain lighting circuits with almost any voltage. Switching to 12- or 24-Vdc systems would eliminate the need for heavy gauge wiring in the ceiling greatly reducing wiring costs, which is particularly high at the moment due to the high price of copper.

Low-voltage circuits would also reduce fire and electrical shock hazards as well, and increase efficiency of photovoltaic/battery systems. Naturally, this switch to low voltage makes more sense for new construction, but it is also feasible for retrofitting office spaces that have dropped tile ceilings. However, a lack of standards and building codes will impede progress here, as will the learning curve for electricians who will be called upon to do the installation.

Long life issues

Other unique challenges face the LED luminaire designer due to one of LEDs best features, their long lifetimes. First, every component of the luminaire must have lifetimes comparable to the LED, which can be on the order of 50,000 to 70,000 hours. LED drivers in particular must be designed for high reliability and long lifetime. Consequently, thermal management is just as important for the LED driver as the LED. Driver component selection is critical, and if you are using a contract manufacturer, you must make sure that your CM doesn’t substitute lower quality parts — a task made even more difficult when your CM is overseas. A distributor with a worldwide presence can help manage your supply chain, ensuring your product is built with the right components.

Even though LEDs are expected to last 10 to 15 years, your customers will expect your luminaires to have an indefinite lifetime. When an LED fails or is too dim after 15 years, the customer will not accept having to buy and install a new luminaire just because the LEDs inside failed. They will expect you, the manufacturer, to provide a replacement LED or LED module.

By the time the LEDs in your luminaire start failing, chances are the LED you initially designed into your luminaire will have gone obsolete 10 years earlier. Without industry standards, your LED board/module design will be unique and there will be no third party making a replacement module or board. Your customers will be very disappointed in you if you do not offer a replacement LED.

It will be fascinating to see how the industry grapples with these bumps in the road over the next few years. As usual, new technologies present both technical and business challenges. While technical issues may appear to be the most difficult hurdles to overcome, often business/regulatory issues impede progress and ultimately determine the outcome of the technology. One thing is for sure — it will be a fun ride. ■