If you pay any attention to backlights - the lighting fixtures that sit behind the LCD panels in your LCD-TV, desktop monitor, notebook PC, digital thermostat, and most other devices using an LCD, and provide the light you see coming through the LCD - you will be aware of (at least) three things:

Ken Werner
Senior Analyst and Editor

• Small displays use LED backlight units (BLUs)
• Nearly all big displays use cold-cathode fluorescent lamp (CCFL) BLUs
• LED BLUs are coming on strong in the mid-size segment, notably notebook PC displays

The discussion concerning large BLUs is how quickly and at what price premium LEDs will make serious inroads on CCFLs turf. At CES, Sharp and Samsung were showing premium LCD-TV lines that use LED backlights, and they looked beautiful. But price will keep these BLUs out of the main stream for some time. Luminus Devices and Global Lighting Technology were showing a 47-inch LCD-TV retrofitted with their edgelit LED BLU, and it looked beautiful. The BLU had been built by Jabil, and Luminus was hopeful that TV sets using the BLU would be commercially available in 2H’08. This technology could significantly reduce the cost of LED backlighting while retaining many (but not all) of the benefits of the "direct" LED backlighting approach being used by Samsung and Sharp.

Eclipsed by the LED excitement are the other alternative BLU technologies we were discussing just a year or two ago: hot-cathode fluorescent lamps (HCFLs), external electrode fluorescent lamps (EEFLs), and flat fluorescent lamps (FFLs).

Philips Lighting was the main champion of HCFLs, and the company heavily promoted its Aptura scanning backlight system, which did a very good job of reducing LCD motion blur. But the high cost of the system prevented Philips from finding customers in the LCD module community, and Philips dropped the product last year.

FFLs generated considerable interest through last year. In March 2006, Chi Mei and Delta formed a JV called NuLight Technology to develop, produce and sell FFLs. And as recently as last year’s CES, Samsung was prominently showing an FFL prototype. To get an idea of what an FFL is, think of a pathway defined by glass barriers between two flat glass plates to form the approximate equivalent of a serpentine CCFL enclosed in a flat structure. Potential benefits were luminance uniformity better than a CCFL BLU and having a one-piece BLU that did not require the module maker to perform expensive assembly operations. The FFLs could also be made without mercury. Although that was touted as a benefit, CCFLs can be made without mercury, too.

All of that sounded good, but a report in today’s Chinese-language Apple Daily, translated by DigiTimes, says that Chi Mei and Delta are thinking of shutting down their JV because high production costs have made volume production impractical. (DigiTimes said the NuLight FFLs were not even cost-competitive with LED backlights, which I find hard to believe.) Chi Mei and Delta declined to comment, but in late November 2007, Chi Mei president Chao-yang Ho acknowledged that production cost was high and production volume was low.

That leaves EEFL, with a simpler internal structure than CCFL and the ability to use one inverter to drive several lamps in a simple way. (Approaches for driving multiple CCFLs from one central power supply have also been developed, but some sources have expressed reservations about excessive complexity. However, work is ongoing, and I hope to have more to say about the state of this technology shortly.)

Reportedly, there were some manufacturing issues with EEFLs for a while, and there now seems to be a distinct cooling of interest among Taiwanese BLU makers. But LG.Philips LCD (LPL) is enthusiastic and plans to introduce an LCD-TV module with 120 Hz frame rate and a scanning backlight using EEFLs this year. We saw this system at an off-site suite at CES last week, and it was impressive, as was the pre-cursor we reported on at FPDI International in Yokohama last October. (In Yokohama, however, LPL was not yet admitting to production plans for 2008.)

Reply to the author