SEDs: Still Standing In The ShadowsCanon and Toshiba's surface-conduction electron-emitting display (SED) technology is taking on cult-like status. But will it ever come to market? 4/29/2006 8:04 AM Eastern
SEDs: Still Standing In The Shadows
Canon and Toshiba's surface-conduction electron-emitting display (SED) technology is taking on cult-like status. But will it ever come to market?
There's no doubt about it. We're in the midst of a revolution in display technology — the likes of which haven't been seen since the first television sets started coming to market in the 1940s. Displays are not only increasing in resolution, they're getting thinner, brighter, and cheaper — all at the same time.
These rapid changes and advances are primarily due to the heated competition between plasma and LCD technologies, as each jockeys to become the anointed successor to the venerable cathode-ray tube (CRT). Manufacturers continue to invest money in larger and larger fabrication lines (fabs, as we call them in the display biz), and have cranked out LCD monitors as large as 100 inches and plasma displays reaching 103 inches.
In fact, there's so much plasma and LCD product in the pipeline that casual observers are figuring this to be the final battle for the flat-panel monitor and TV space. Yes, there are other options, such as organic light-emitting diodes (OLEDs), but that technology is difficult to produce in large screen sizes, and its advocates are focusing more on the low-hanging and plentiful fruit of the handheld and near-to-eye display market segment.
As inexpensive as plasma and LCD displays have become, they still have shortcomings when compared to the performance of a CRT, which is known for its smooth, linear grayscales, accurate color rendering, and wide dynamic range. Even the best plasma monitors have trouble matching the full color gamut of a CRT, and require a lot of sub-field processing to achieve low black levels and smooth grayscales.
As for LCD imaging, it has improved tremendously in the past five years, but still exhibits higher black levels and comes up short in colorimetry when using conventional illuminants, such as cold-cathode fluorescent lamps (CCFLs). It also has problems with fast motion smear. All of these problems are being addressed rather aggressively, as you might expect. There's simply too much money being chased in this market, particularly with respect to consumer TV sets. With more Gen 6, 7, and 8 LCD and plasma fabs coming on-line, all of the players in this high-stakes poker game continue to ante up.
Now you see it, now you don't
Crowning LCD or plasma as the new king isn't quite a fait accompli. Lurking in the shadows is yet another contender — one that potentially could knock both plasma and LCD off the pedestal as the best-looking flat-panel display ever made. It's been in development for nearly 20 years, but has ventured out in public only sporadically. Just like Elvis used to do, it has also made a couple of unannounced appearances in Las Vegas, dazzling those who were fortunate enough to see it and further building up its myth.
This shadowy contender goes by the name SED, which is shorthand for Surface-conduction Electron-emitting Display (you can see why they stuck with SED). Technically speaking, the SED falls into a category of displays known as FEDs, or Field Emission Displays. Practically speaking, the SED is a close cousin to the CRT, except on a real strict diet. Whereas a typical CRT might measure 16 to 18 inches in depth, the SED takes up a couple of inches. But it shares other qualities with CRTs, such as its use of high-voltage anode potential with low current, low-voltage scanning cathode electrodes, and color phosphor formulations for red, green, and blue channels.
If you've been lucky enough to see an SED demonstration, you probably were impressed by several things. First, the deep black levels resemble that of a CRT, and there's plenty of contrast “punch” in the image. Next, the colors are amazingly life-like — not over- or undersaturated. There's lots of dynamic range, too. Images have plenty of detail from deep shadows to near white. And don't forget about the real-time wattmeter. At CES 2005, the demo I attended showed power consumption to be in the range of 80 to 100 watts for a 36-inch display. That's almost one-fourth of what some 42-inch plasma monitors draw and less than half the power demand of same-size LCD monitors using CCFLs.
Behind the glass
Like every other magic show in Vegas, there are plenty of “how'd they do it?” questions associated with the SED, so I'll discuss what I know from publicly released information and the few presentations made at Society for Information Display (SID) trade shows by representatives of SED Inc., a 2004 joint venture between Canon and Toshiba.
First off, the SED retains its bloodlines to the CRT by staying with the high-voltage emissive imaging process. One of the reasons for such beautiful pictures on a CRT is its high-voltage, low-current operation, mentioned earlier. Small changes in the cathode current of the tube effect linear, large changes in anode current — an attribute also favored by fans of tube-type musical instrument amplifiers and home audio systems.
Because of the CRT's linear operation, it can reproduce a wide range of gray levels or shades, which results in an equally wide range of color shades. The CRT can be driven to high levels of peak brightness in small areas of the image, but it's also equally at home with low luminance levels and shadow detail. The only limiting factor of the CRT is that picture resolution and image brightness are forever linked — the scanning electron beam's intensity limits its resolving power. Drive the beam harder for higher brightness and the spot size increases, thus decreasing resolution.
That link, which is broken in LCD and plasma technology, is also severed in the SED. A backplane of scanning electrodes (cathodes) drives individual red, green, and blue pixels in a stripe configuration. The path of emission is straightaway from cathode to anode; there's no need for deflection yokes and magnetic fields to pull the electron beam back and forth.
In theory, the resolution of an SED can be as high as anything we've seen come out of a plasma or LCD fab; perhaps even higher. At SID 2005, SED Inc. presented two papers that spilled the beans a bit more on the 36-inch prototype that was making occasional appearances at trade shows. It had 1280x720 resolution, a response time of less than 1 millisecond, 10,000:1 contrast, and peak brightness of 400 nits.
We've seen some or all of those numbers attributed previously to plasma and LCD technologies, but never all in one package — not one where we could see the demonstration and feel the numbers were real and not hype.
The process of forming SED scanning electrodes was also covered at the conference. While beyond the scope of this article, suffice it to say that the fab process involves inkjet printing a 10-nanometer layer of palladium oxide (PdO) and forming nanogaps in this layer after applying a second carbon film, creating the numerous cathode electrodes. More specifics weren't forthcoming, except to say that image brightness decreased by only 10 percent over a 60,000-hour accelerated life test.
Too good to be true?
Okay, stop drooling. Yes, the SED does sound as if it's too good to be true, and it has certainly taken Canon a long time to get any kind of prototype to market. The formation of SED Inc. in 2004 was certainly an acknowledgement by Canon that no matter how good its engineering and technical prowess, it would have a difficult time manufacturing and mass-marketing this technology on its own.
While CES 2005 was the moment for SED to prove its technology was alive and kicking, CEATAC 2005 and CES 2006 showed that SED Inc. could make multiple versions of that same 36-inch display with repeatable image quality and consistency. Although an impressive demonstration, it disappointed many of us expecting a 50-inch or larger 1920x1080 prototype — hinted at by SED Inc. executives a year ago.
Mixed signals with SED demos are nothing new. At the CES 2005 screening, SED representatives talked about rolling out the “Ferrari” of flat-panel displays and going after the high end of the market. Trouble is, the high end of the market isn't where the rest of the flat-panel display marketers are focused these days. They're building bigger fabs and ramping up production to capture market share as the LCD-plasma wars heat up.
So, the real problem that SED Inc. faces (and one that could prevent this technology from ever achieving its potential) is jumping into an already-crowded field of flat-panel giants and wannabes, trying to capture any market share. Selling to the high end is a good strategy for a few months, but the economics of flat-panel display manufacturing demand high-volume production out of the gate to get yields up and costs down quickly. That means moving your flat-panel product through mainstream sales channels and fighting market share battles against the established LCD and plasma goliaths like Matsushita, Sharp, LG, and Samsung — all companies who make tens if not hundreds of thousands of LCD and/or plasma panels every month. And there's more competition coming from Asian tigers AU Optronics and Chi Mei Optoelectronics, two fast-growing Taiwanese powerhouses in LCD manufacturing.
A rough ride
In my display technology seminars, I have on more than one occasion described the SED as “the best display technology you've ever seen that may be stillborn.” That comment may seem a bit harsh, but consider these facts.
Pricing wars. Back in 1986, when Canon set about its initial research, we lived in a world where high-quality display technologies would command and realize five- and six-figure prices at retail. Even as recently as six years ago, you could ask for and get $10,000 to $15,000 for a 50-inch plasma display.
Today, market focus has changed to high volume shipments and “take-no-prisoners” battles for market share. With 50-inch plasma TVs now selling for about $3,000 and 37-inch plasma monitors going for $2,500 and less at retail, it doesn't take a doctorate degree in economics to see that an unproven technology such as the SED has a tough uphill climb ahead of it.
Fighting the odds. SED Inc. may be prepared to spend hundreds of millions of dollars to make the SED viable (the joint venture was capitalized with $1 billion yen a year and a half ago), but it must quickly manufacture tens of thousands of finished, functioning SED monitors and/or TVs each month at competitive prices just to have a shot at shelf space in the big box stores. That “sink or swim” business model is scary to venture capitalists.
No one is king of the hill forever. With the widespread availability of LCD and plasma TVs and monitors in sizes from 15 to 65 inches, new fabs coming on-line every few months that use larger sheets of motherglass, and a steady decrease in price across screen sizes up to 65 inches, the “holes” in the flat-panel TV and monitor marketplace are getting filled in short order.
About the only edge the SED still has is in power consumption, but that could change — pulsed backlights for LCD displays promise not only improved color, but also substantial reductions in power drain — an issue even the plasma folks are still wrestling with.
The history of display technology is littered with “best image quality” solutions that never realized their potential in the marketplace, from Hughes-JVC's Image Light Amplifier (too unstable) to laser/CRT projectors (too bulky and underpowered in the blue channel) and the grating light valve (obviated by Sony's 4K SXRD technology).
Is the SED destined to join that group, or will it eventually flourish?
Pete Putman is a contributing editor for Pro AV and president of ROAM Consulting, Doylestown, PA. Especially well known for the product testing/development services he provides manufacturers of projectors, monitors, integrated TVs, and display interfaces, he has also authored hundreds of technical articles, reviews, and columns for industry trade and consumer magazines over the last two decades. You can reach him at email@example.com.