Touchscreens have been available since the days of cathode-ray tubes, but the technology didn’t really catch on with consumers until mobile phone makers adopted it to solve the tiny-button problem. Now touchscreen smartphones and tablets collectively constitute the fastest-growing electronics market segment.
According to DisplaySearch (Santa Clara, California), shipments of touchscreen tablets are forecast to reach 60 million units in 2011 and could top 260 million units by 2016. Add to that the more than 400 million mobile phone touchscreens predicted by IHS iSuppli Corp. (El Segundo, California), and the total market could top $10 billion this year (see sidebar, at the end of this article).
“Touchscreens have been around for a long time, but they were only popular in business and industrial settings, such as food service, airport kiosks and industrial keypads,” said Rhoda Alexander, director of monitor research at IHS iSuppli. “The real transition for consumers … was when Apple moved into smartphones
and then tablets. Before then, consumer touchscreens didn’t work very well, because they had to operate a standard OS. But with the move to smartphones and tablets, operating systems like iOS have enabled a very touch-friendly user interface.”
Google’s Android OS—the first major competitor to Apple’s iOS — did not support multitouch at introduction, but the latest incarnation accommodates a wide array of multitouch gestures. Some of them—including spin, thrust and slice—are unique to Android; all will work identically on any Android smartphone or touchscreen tablet. The BlackBerry Tablet OS and Windows Phone OS have
similarly become touch-enabled.
“Handset makers used to give us a long list of obstacles to adopting touchscreens, but when Apple introduced the iPhone all those obstacles suddenly seemed surmountable,” said Andrew Hsu, technology strategist at Synaptics Inc.(Santa Clara, California). Synaptics began as an evangelist for the benefits of touchpads as a substitute for a PC mouse but has since reinvented itself as a touchscreen controllersupplier for mobile handsets. It claims a number of major design wins,including one in Google’s Nexus One smartphone.
Just a handful of contrary trends threaten the touchscreen industry. Foremost are competing technologies that deliver a similar user experience without the expensive touchscreen hardware, such as the 3-D gesture recognition made possible by Microsoft’s Kinect gaming interface, which uses cameras and pattern
recognition to sidestep the need for the sensors required by handheld controllers. Kinect-like 3-D gesture recognition, using infrared rangefinder technology Microsoft acquired from Canesta, is being downsized for Windows phones and tablets. Armed with a touch-enabled version of Windows that works across all device sizes—from its own 40-inch Surface to its licensed touchscreen tablets and
smartphones—Microsoft could redefine the touchscreen landscape.
Meanwhile, all the LCD makers are retooling their manufacturing lines to incorporate touchscreen sensors directly into their displays, a move that would eliminate the need for today’s OEM add-ons. Samsung and Nokia, for instance, have already integrated touchscreens into organic LED displays for their respective Galaxy S and N8 smartphones.
Alternative materials for integrated touchscreen sensors are also on the horizon, including Cambrios Technologies’ ClearOhm transparent conductors, using silver
nanowires; C3Nano’s carbon nanotube films; 3M’s copper-mesh films; repurposed polyethylenedioxythiophene conductive polymers; and epitaxial graphene films from a variety of vendors. All aim to slash the cost of touch sensors over the increasingly rare indium tin oxide (ITO) used for touch sensors today.
Touchscreens consist of a transparent sensor layer attached directly to the
controller chip and sandwiched between a top glass cover and the display on the bottom
Typical layer stacks for resistive (left) and capacitive (right) screens
Touchscreens’ runaway market appeal has given rise to a mature worldwide supply chain in just a few years, encompassing the manufacturing plants in Taiwan
and Japan where the transparent sensors are fabricated; the U.S. and European manufacturers of the controller chips that translate changes in resistance or capacitance into finger-down locations; and module makers and system integrators that add the clear cover, laminate the transparent films and integrate the electronics.
Legacy, resistive touchscreen technology uses two conductive polymers on separate layers that can be deformed to touch each other wherever a finger or stylus touches the top layer. Resistive controllers—which are available off the shelf from Analog Devices, Texas Instruments, STMicroelectronics and other
mixed-signal chip makers—are relatively simple and very accurate, but they do not usually recognize multiple touches. Divergent architectures have been developed for resistive touchscreens, using varying numbers of connecting wires (such as four-wire or eight-wire) to make the task simpler or more accurate for specific applications.
“Touchscreen technology is extremely diverse, with many methods specialized for different applications; but in general resistive is the legacy technology, while
projected capacitive has recently become the industry leader,” said Jenny Colegrove, vice president for emerging display technologies at DisplaySearch.
Projected capacitive touchscreens rule the roost in high-end mobile devices. Smart appliances and security keypads have no real need for capacitive touchscreens, however, and for some applications even resistive touchscreens are overkill. “Resistive touchscreens are still popular because of [the technology’s] maturity and low price point, whereas projected capacitive still has some issues
with yields and the lamination process for large-scale screens,” said IHS iSuppli’s Alexander. “As long as you have a variety of screen sizes, application environments, price points and use cases for these devices you are still going to need a variety of touchscreen technologies available.”
Projected capacitive technology drives one plate of a transparent capacitor with a signal, then measures the results at the adjacent plate with an analog-to-digital converter. The capacitive sensors are usually cast in a diamond pattern, with one
diamond-shaped capacitor plate on each side of the glass or both on the same side, using whisker-sized jumpers. A smartphone uses a couple of hundred capacitive sensors and tablets up to 10 times as many, making it possible for a smart controller chip to discern any number of touches at resolutions fine enough to detect even the smallest child’s finger. Several controller makers, including Cypress and Integrated Device Technology (IDT), are proposing proprietary patterns that eliminate the need for jumpers.
Multitouch gesture capability began with two fingers to zoom, three to scroll and four to swipe; it’s now become a free-for-all, as multitouch variations proliferate to enable finer manipulation of on-screen objects. High-quality transparent sensor
patterns support smart gesture recognition, but the smarts originate in controller algorithms that debounce and condition the signal from multiple fingers. The touchscreen controller sends the information to the application processor, which in turn identifies gestures of varying complexity, such as tap- to-select, brush-to-scroll and pinch-to-zoom.
Touchscreens are evolving toward allowing multiple finger touches to manipulate objects on-screen in much the same way objects might be handled on a real desktop
Taiwan and Japan today manufacture most of the world’s high-end projected
capacitive touchscreens. Taiwan’s Wintek and TPK Touch Systems (both Apple suppliers), along with faster-growing siblings such as Young Fast and J-Touch (both also based Taiwan), account for the majority of worldwide touchscreen shipments. Japanese suppliers Gunze, Suzutora and DMC (located in Osaka but partnered with Austin-based Touch International) are quickly ramping up competing manufacturing operations, as are three dozen other touchscreen makers worldwide.
Chinese manufacturers are not yet making high-end capacitive multitouchscreens, which are difficult to use with a stylus, but have stuck with resistive touchscreens, which are better at rendering Chinese characters.
Touchscreen controller chips come from such suppliers as Atmel (San Jose, California), which supplies Motorola’s Xoom, Samsung’s Galaxy Tab and many other Android-based tablets and smartphones; Cypress Semiconductor (San Jose), which supplies RIM’s BlackBerry PlayBook, the Barnes and Noble Nook and dozens of smartphones, including HTC models; Synaptics, which supplies the
IDEOS S7 Slim tablet as well as many other Android and Windows phones;
Broadcom (Irvine, California), whose touch controllers are found in many iOS devices, including the iPad 2; and Texas Instruments (Dallas), whose touchscreen controllers are present in all iPhone models, including the iPhone 4, according to Chipworks.
Atmel takes the design win crown, having landed slots not only in the Xoom and Galaxy Tab but also in LG’s G-Slate, Acer’s Iconia, Asus’ Transformer, Dell’s Streak and nearly every other Android-based tablet, plus eight out of 10 of the world’s top smartphones, from HTC’s 4G Evo to Motorola’s Droid.
“Atmel has the fastest response time by a mile — 300 compared with 70 samples/second — which means gestures are detected more accurately,” said Sherif Hanna, Atmel product marketing manager. “We also have superfast first-touch latency of 8 to 12 milliseconds—twice as fast as our competitors.”
Atmel’s main competition today comes from Cypress and Synaptics, but the field is about to get more crowded, since nearly every other semiconductor maker with mixed-signal capabilities wants in on the action. Industrial giant STMicroelectronics added a projected capacitive controller chip to its S-Touch line last fall and will ship the device this summer. IDT this year announced its
proprietary PureTouch technology, promising to lower the cost of capacitive touchscreens by using a single layer of sensors rather than the two sensor layers of expensive ITO required today. Cypress, too, is pitching a single-sensor-layer solution with its TrueTouch technology.
Silicon Labs, which supplies the microcontroller in NextWindow’s optical touchscreens, is gearing up for projected capacitive touchscreens by creating a microcontroller that leverages SiLabs’ know-how in self-capacitance technology for buttons.
“Right now our controllers are recognizing self-capacitance on individual buttons by measuring the capacitance of one plate in relation to earth ground,” said Steve
Gerber, director of human interface products at Silicon Labs. “For projected capacitive touchscreens, however, we need to measure the mutual-capacitance between two elements, which for us means designing a specialized A/D converter peripheral, very similar to the self-capacitance block we already have on our 8051-based microcontrollers.” The company is currently sampling an 8051 with an integrated projected capacitive touchscreen controller.
Freescale also has surface capacitive touchpad blocks on many of its microcontrollers, including its newest line of ARM-based Kinetis MCUs, and has resistive touchscreen solutions based on its S08, ColdFire+ and i.MX processors. Freescale has not made any announcements about projected capacitive controllers but is likely to begin offering them as peripheral blocks on its
microcontrollers by 2012.
Analog Devices offers a family of controller chips for low-cost resistive touchscreens, making its touch controllers popular in applications ranging from point-of-sale terminals to smartphones. ADI also has expertise in capacitive touch technology with its high-precision capacitance-to-digital converter (CDC), used for proximity sensing. Like Freescale and Silicon Labs, ADI has not yet announced entry into the controller market for projected capacitive touchscreens, but its expertise in noise management and its high-performance CDC portend a projected capacitive touchscreen controller announcement, probably in 2012.
In a capacitive touchscreen, the signal intensity levels for the rows and columns denote the touch location (which changes their capacitance)
For the future, more than a dozen competing touchscreen technologies — from
acoustic wave to near-field imaging—promise to sidestep the quirks of projected capacitive touchscreens for specialized input needs, such as those of glove-wearing medical professionals. Among the companies vowing to provide multiple types of touchscreens to meet the various vertical market requirements are Elo Touchsystems (Tyco Electronics) and 3M.
One of the most promising alternatives is optical touch technology, which uses ultralow-power infrared LEDs and photodetectors to pinpoint gestures without requiring glass or ITO. NextWindow, for instance, supplies the touchscreen
capability in the bezel around Hewlett-Packard’s TouchSmart PCs. Similarly, TI has partnered with Neonode Inc. (Stockholm) to shrink the optical bezel profile to 1 mm high for mobile device touchscreens.
Infrared LEDs feed photodetectors in a bezel around the screen edges, sending beams of light across the top of the touchscreen that when interrupted can be used to detect finger touches (SOURCE: Neonode)
TI has been in the touchscreen controller business since acquiring Burr-Brown a decade ago. Recently, TI announced special models of its ultralow-power MSP-430 microcontrollers with built-in support for both resistive touchscreens
and surface capacitive touchpads. Now its collaboration with Neonode propels it into optical touchscreens. “Neonode’s implementation of optical touchscreens is almost a hybrid of existing resistive and capacitive touchscreen solutions, said Adrian Valenzuela, team lead for TI’s touch portfolio. “You get the form factor and response of capacitive—only a light touch is required, and gestures are easy to recognize—but it’s a lot more cost-effective, like a resistive display.”
Neonode’s first major design win for its zForce patented optical touchscreen came late last year in the Sony Reader. Kobo (Toronto) last month picked zForce for its eReader.
“Our technology is particularly well suited for e-readers, because zForce only consumes microamps of power in-between touch events,” said Thomas Eriksson, chief executive officer at Neonode. “Finger touches are detected whenever they break a beam, which uses much less power than either resistive or capacitive
touchscreens. Also, since there are no overlays on the screen itself, zForce doesn’t dim the light reflected from E-Ink’s and Qualcomm’s e-reader displays.”
Chinese e-reader maker Hanvon recently announced availability of its own electromagnetic resonance touchscreen. Hanvon claims OEMs choosing its proprietary dual-touch ERT need only integrate antenna sensors on their circuit board to achieve the accuracy of resistive technology for stylus input of Chinese
characters, while obtaining multitouch capabilities usually found only on projected capacitive touchscreens.
Sidebar: Touch extends its reach
“One out of three mobile phones uses a touchpanel in 2011,” said Jae Shin,
marketing director of market research firm Displaybank. That will increase to one out of every two phones by 2014, the market research firm projects.
The touchpanel market this year will log $10.42 billion in revenues, for 76 percent growth year on year. Panels less than 10 inches will account for 89 percent of total 2011 revenues. Projected capacitive touch technology is expected to account for 73 percent of the total touchpanel market in 2014.
At the Society for Information Display’s Future of Touch conference last month, experts looked at what the next phases of touch technology will bring to various markets. John F. Jacobs, customer value chain manager at Cisco Systems, suggested that developers ask the following questions when weighing the relative
merits of the various touchscreen approaches and technologies for their applications:
• What are the trade-offs between multitouch and so-called touch 1.5 (pinch-to-zoom) technologies?
• Do you really need “true” multitouch?
• Is the feature set clearly defined early in the development process?
• How does the marketing “wish list” measure up against the user requirements?
• What are the “ideal” display size and resolution for the device?
• How will the device be held? One hand or two? Cradled or gripped?
• How will it be operated? With fingers? Thumbs?
If one does not have ready answers to those questions, then just providing touch capability for its own sake in the product is not a worthy goal, according to Jacobs.
Bob Mackey, principal scientist at Synaptics, heeded the warning that the integration of touch and the display hardware is of utmost importance. “You can’t just have a great capacitive touch technology and not have an integration plan that takes all elements into account,” said Mackey. “Remember, to a systems
engineer, everything is part of a system.”
Toward that end, Mackey listed the perceived shortcomings of transparent conductors for capacitive touch sensing. “While volumes are increasing, pushing costs down, capacitive touch sensing does require transparent conductors—which often aren’t so transparent, and not very conductive,” he said.
The required indium tin oxide (ITO) layer used in capacitive touch sensing may be expensive, but it fits the existing supply chain and offers invisible patterns when used with good index matching, said Mackey. No effective volume replacement for ITO has been found, he added. For the future, however, Mackey looks to
structured micro- or nanomaterials to offer a better combination of optical transparency and electrical conductance than ITO provides.
Vernon Spencer, founder and managing director of Visual Planet, offered his own to-do list for moving touch technologies forward. Because touchpanel providers are pursuing divergent approaches, Spencer cited the need for such basics as
educating the developer community, arriving at common touch standards, agreeing on definitions for multitouch and providing software development kits. “Hardware needs to coexist, and a high-profile multiuser productivity tool needs to be developed,” said Spencer.
Visual Planet specializes in the manufacture of large (30- to 167-inch), flexible projected capacitive touchscreen foils, which Spencer said “work through any nonmetallic surface and create a fully functional touchscreen.”
In a SID conference paper, researchers from FlatFrog Laboratories AB (Lund, Sweden) cited drawbacks of projected capacitive systems. For one, the technology is highly dependent on the availability of indium, a rare and increasingly scarce element that is a key component of the ITO layer in touchpanels. The
supply issues for indium will continue to inflate the bill of materials for projected capacitive touch systems, the researchers predicted. In addition, the inherent properties of the ITO layer itself impose size limitations and pose production yield issues that hinder development of cost-efficient projected capacitive multitouch systems at larger sizes, according to the paper.
The researchers said FlatFrog’s planar scatter detection (PSD) approach maintains the advantages of multitouch systems without imposing the drawbacks normally associated with projected capacitive or other optical touch systems. PSD applies frustrated total internal reflection (FTIR) in combination with
proprietary optical detection and advanced decoding algorithms. — Nicolas Mokhoff