Let's talk about what's important in solid-state lighting technology, and what trends are mere sideshows.
Talk is cheap when one is not well-informed. Here I'll try to put a few LED topics in perspective. What's the difference between what excites the blogosphere and where technology is actually getting traction?
Where it's at.
Indication and illumination
Lighting has always involved these two main categories. LEDs are not different from fire in that regard. For over 40 years, LED growth has been mainly in indication applications: panel lights, traffic lights, neon-replacement signs, graphic displays, etc. Phone back-lighting falls into that same category -- with acknowledgement you can use a phone as a five-lumen flashlight. Now that mainstream white-light illumination is here, we are debating all the aspects of that. Let's leave flashlights out of it for now.
LED indication wants to tell us something. With indicators, it's all about longevity, miniaturization, digital/software controllability, and reduced power consumption. On the other hand, LED illumination wants us to see something better.
But there is one more sub-category of illumination that, while valid, tends to get blog attention all out of proportion to its contribution to industry growth… and that is appearance.
This is my term for architectural lighting, where the purpose is to enhance the visuals. Cost, size, and efficiency are secondary. Here, art is mixed with science. Unfortunately, we often read about developments which are fascinating but do not necessarily influence future trends in mainstream commercial lighting. OLED and all its derivatives are that.
A cousin of the film-based electroluminescent panels in vogue 40 years ago for aircraft cockpits and exit signs, OLED technology is more expensive to produce, less efficient at converting electricity to light, and incapable of being usefully collimated (a big deal in mainstream applications). However, OLEDs will often be a choice for aesthetics where special effects or unusual backlighting form factors are needed. But don't confuse those OLED features with being the next big thing for general illumination.
Talk is cheap
Several technology predictions that have stuck in my mind over the years
1975 -- David Sarnoff, RCA chairman, predicted infant microprocessor technology would become ubiquitous in every kind of consumer product. "Right," I said to myself. Turns out that prediction was 100 percent accurate, and it's happening far faster than anyone envisioned.
1982 -- Ken Olsen, founder and CEO of Digital Equipment Corp., famously asked, "Why would anybody need a computer in the home?"
1985 -- I participated in conference on "smartpower" IC's (the rage at that time) where a GE executive, supported by barrage of media publicity, predicted that in a decade most major home appliances would have embedded smart chips... controlled from a central controller. Oops! -- didn't happen, and GE soon got out of the smart-chip business.
2004 -- I was invited to sit in with a conference room full of salesmen. A Dialight sales manager said, "See those fluorescent tubes up above? They will all be replaced by LED versions in four to five years." "Right," I said to myself. Didn't happen, and it's not likely to happen in the next 10 years either.
Contrary to the many anti-CFL opinions we read in the blogosphere, CFLs are jumping off the shelves in hardware stores across the US. Hundreds of millions have been sold, and the trend is up. The timeworn comments about the inadequacy of CFL color do not hold much water any more. A 2700K-3000K bulb, selling for less than $2, will light an area in ways virtually indistinguishable from a 60-watt incandescent for 90 percent of sockets. CFLs are available in several color temperatures for those who do not want warm white -- a choice not readily available in incandescent A19s, and not universally so in LEDs.
For those who spend too much time contemplating CRI purity (for mainstream applications), try putting some crayons near a new CFL bulb. You'll be surprised at the color rendition, even the red.
Do CFLs really give low CRI and R9 values? They're actually not all that bad: the red will look red unless you nitpick. CFLs with wavelength spikes at blue, green, and red can still do a pretty good job of fooling eyeballs. It has been shown that humans react very favorably to saturated colors. Even though overall CRI may be well below 80-85, in a supermarket produce section a CFL with good blue, green, and especially R9 (red), can make veggie colors "pop," as the retail lighting-industry vernacular goes.
A CFL can stand up to use in an air-restricted space far better than an A19 LED bulb. Why would folks buy 60-watt equivalent LED bulbs in droves? Even when and if the price gets to $2, there will be CFLs for under $1 that will be more heat-forgiving if installed in constrained sockets. So don't hold your breath just yet for widespread use of A19 LED bulbs.
On the other hand, for recessed down-lights, LED PAR lamps will be the bulbs of choice within a few years. An LED PAR lamp will deliver two or three times more light to a surface than a same-wattage PAR-type CFL. You'd be astounded by the poor performance CFL PAR lamps show directing light -- if you went to the trouble of doing serious testing.
Where the real action is
It's happening primarily in no-nonsense canopy lighting, high-bay (heights of 30-40 ft., 9-12 m) and track lighting in hundreds of thousands of warehouses, factories, large-mall retailers, supermarkets, museums, restaurants, etc., across the US and Canada.
Big-box retailers and home-improvement centers have no incentive to switch from their very cheap 8-foot T5 fluorescent "low-bay" (15-20 ft., 4-6 m) fixtures.
The real growth is in two areas: first, the large number of PAR-lamp, commercial applications where usage is 18 hours/day -- that scenario is now well past the ROI tipping point -- and second, in high-bay, large-area lighting. Traditionally, folks made direct lumen-per-Watt comparisons among fluorescent, HID, and LEDs. That's changing, with more awareness that "delivered" lumen patterns from LED luminaires can be shaped, in ways impossible with other light sources, so that minimum foot-candles on the floor can be 40 percent higher than for luminaires of similar lumen output -- a 40 percent gain in energy-efficiency!
Not as much fun to talk about as color-changing bulbs or OLED panels that can be formed into $1,000 giant, glowing flower petals. Nevertheless these are the horses that are pulling the industry's wagon.