With consumers clamoring for the next generation of mobile phones, manufacturers must figure out how to reduce the cost and complexity of today’s touchscreen technology. For mobile phones to evolve, so must their core technologies, including touchscreens. This article explains key touchscreen performance parameters, critical touchscreen manufacturing processes and their impact, significant physical design tradeoffs, and new discoveries in advanced capacitive touchscreen technologies including single-layer sensors, and in-cell and on-cell technology.
Despite the fact that millions of users enjoy the benefits of capacitive touchscreen enabled phones, few users understand the underlying technology. A capacitive touchscreen, like those used in iPhone and Samsung Galaxy products, is commonly constructed of several layers of materials. This layering construction is called the 'stackup' (Figure 1 ).
In mobile products, the top layer is a protective layer composed of glass with an anti-scratch coating, or PMMA (polymethyl methacrylate), commonly called plexiglas or acrylic. This top layer is often shaped, back-printed with the company logo or decorative graphics and button indicators, and is the outward-most facing material of the touch product. Directly underneath the surface layer is a layer of thin adhesive and then the electrically conductive layers for touch sensing.
Most touch sensors are built using a combination of layers of glass or acrylic, isolation layers, clear adhesives, and Indium Tin Oxide (ITO). ITO is a ceramic-like material known for its high conductivity and excellent transparency. While ITO is broadly used and has been proven to be an excellent material for touchscreens, handling and manufacturing ITO has its disadvantages. The primary objection is that ITO is expensive, the materials are fragile and heavy, and the manufacturing process is labor intensive and expensive.
Basics of ITO stack sensor fabrication
The manufacturing of an ITO sensor is similar to printing a complicated multi-color poster or graphic where every individual color must be printed in a separate pass. Much like a sophisticated printing process, ITO is deposited on the sensor substrate in multiple layers to attain the desired electrical pattern.
Figure 2 shows a typical process flow for manufacturing ITO-based sensors. Steps include sputtering ITO powder over glass, thermal baking the ITO to its melting point thereby creating a conductive layer, and then etching the sensing circuit topology on the conductive layer by use of photo or laser lithography. Every manufacturing step adds cost as a result of materials cost, manufacturing time, and yield loss (every step risks further defects).
Click on image to enlarge.
Traditional capacitive touchscreens use 1, 2, or even 3 layers of ITO depending on the specific product design considerations and the touch panel supplier technical capabilities. Unfortunately, every required layer of material adds costs, thickness, and weight.
In the case of mobile phone touchscreens, the thickness can depend heavily on the materials used (Figure 3 ). For example, a typical glass coverlens varies from 0.5-1.0 mm. Typical PMMA lenses, though lighter, are usually 1.0 mm thick or greater.
Click on image to enlarge.
The primary reason is that PMMA is a more flexible material than glass. If the PMMA is not thick enough, it can actually flex from finger pressure, causing mechanical and electrical problems as the ITO materials in the stackup get closer or even touch the LCD. A sensor’s secondary substrate of glass will generally have a thickness of 0.2 -0.7 mm while a similar structure of PET (polyethylene terephthalate) has a thickness of only .055 mm.
However, PET cannot yet easily be printed with bridges or jumpers, so multiple layers are needed as opposed to a glass substrate that better handles ITO etching and bridges. Adding multiple extra manufacturing steps for PET must then be considered against the higher materials cost, thickness, and weight of glass.
There are still many mobile products today that are created using an airgap (see third example in Figure 3 ).This airgap is used for both commercial and electrical reasons. First,most mobile manufacturers like to dual-source their LCDs and sensorsupply to manage supply chain shortages and costs. If they elect to bondthe sensor to the LCD, they could limit their supply chain optionsbecause some suppliers are not capable of direct bonding in house.Adding an airgap creates a supply chain where virtually any reputableContract Manufacturer (CM) can add a gasket bond to the edge of a sensorand bond it to an LCD of choice.
Technically, LCDs emit asubstantial amount of electrical noise. Because capacitive touch sensorsmeasure minute changes in capacitive signature on a touchpanel, thiscapacitive field can be disrupted by LCD switching noise. In the case ofa typical TFT LCD, the common electrode (VCOM pictured in Figure 5) isdriven by either a DC or an AC voltage.
This DC or AC voltagecan switch dramatically as the LCD changes patterns on the screen. As aresult, this switching noise can appear to be a false finger touch.Adding an airgap of 0.2 mm or greater avoids this direct noise injectioninto the touchpanel. While an airgap can be beneficial electrically, itis a large cost to pay in the thickness of a mobile product. Airgapsalso create a visual parallax, which is the difference between theperceived position of an object on the touchpanel and its actualposition on the LCD screen.
This space creates a differencebetween perception and reality. If a user is viewing the product’sscreen at an angle, the human eye may believe the finger should beplaced in one location when in fact it is not really over the desiredlocation on the LCD. This problem of parallax is a major one for smoothuser interface operation that mobile device manufacturers wish toavoid…if they can afford to.
As a result, manufacturers havedeveloped dozens of permutations of layering in an effort to drive downcomplexity, eliminate parallax, manage noise injection, and maintaincontrol over their supply chain. Some vendors are attempting to do awaywith several layers of material by printing only on the touchscreencover lens. The industry is calling this 'one glass', 'direct patternwindow', 'sensor on lens', or 'G1' (for 'one glass layer').
Indevelopment of a sensor on lens solution, the sheet glass is firststrengthened (hardened through chemical treatment), has ITO depositedand etched, and then is cut into smaller pieces for use in products.Finally, an advanced post-process technology is applied to strengthenthe glass edge to prevent edge-break susceptibility. Alternatively, somemanufacturers cut larger glass into smaller pieces, chemicallystrengthen them prior to ITO deposition, and then apply the ITOprocessing.
Figure 4 shows many variations of actualmanufacturer stackup options, the most sophisticated of which are thesensor on lens, on-cell, and In-cell.
Click on image to enlarge.
SLIMing down to lower costs
Tocreate sensors with the least amount of material and processing, touchmanufacturers must develop intellectual property that allows for highperformance (speed, accuracy, and multitouch capability) while using theleast number of manufacturing steps. As an example, to do this Cypress Semiconductor's SLIM (Single Layer Independent Multitouch) technology uses a unique combination of ITO patterning and touch controlleralgorithms that allow a sensor to be built with a single ITO layer thatdoes not require vias (bridges or jumpers).
As a result,multiple steps of photoresist patterning and removal can be eliminatedand multiple layers of glass, adhesive, and acrylic can be removed. Allof this makes for a simple, cost and weight efficient sensor. As aresult of process step removal, the approach can cut the cost of amultitouch panel in half. In addition to SLIM, some companies arepursuing solutions to eliminate the upper sensor layer altogether bydesigning the ITO transmit-and-receive layers directly into the LCD,known as on-cell and in-cell touch technologies.
The basicpromise of on-cell and in-cell is to eliminate the entire separatesensor development. In on-cell, the ITO sense layers are patterneddirectly underneath the top polarizer layer in an LCD. LCD vendors will,in effect, form the touchscreen layers for transmit-and-receivedirectly into the LCD glass substrate.
Because many glasssensors are between 1.0-1.5 mm thick, this technology can eliminatesubstantial thickness and weight in a mobile product. For in-cell, thepromise is for the ITO pattern to be manufactured simultaneously withthe VCOM layer of an LCD. This pushes the touch sensing channels deepinto LCD. The differences in these two technologies can be seen in Figure 5 .
Click on image to enlarge.
However,the reality of in-cell and on-cell is that there are significantcommercial and technical barriers for these technologies to becomeubiquitous. For on-cell, the primary issue is LCD injected noise. Thisis a significant technical barrier, as the noise profile of the LCD mustbe managed by the touch controller analog sense channel and filteringalgorithms. As the sensing ITO is designed closer and closer to the TFTswitching elements, this noise grows substantially.
For thisreason, on-cell technology is reserved for high end AMOLED (activematrix OLED) that produce few LCD noise spikes, or for low noise profileLCDs. Unfortunately, this puts on-cell technology out of reach for manylower-end mobile phone and consumer touch products.
In the caseof in-cell, the ITO sensing elements are manufactured as part of theactual VCOM layer, thereby putting the sense elements directly withinthe LCD noise source. Again, there are only a few high-end solutionsthat use In-cell today. While they have eliminated the use of an uppersensor, they are manufactured by few companies, are limited in supply,and are expensive. While it is conceivable that more LCDs will comeequipped with ITO sense capability in the future, the companies that areexpert in LCD and touch sensing technologies are not always the same.As a result, making on-cell and in-cell more common will takesubstantial partnership effort to combine or merge intellectualproperty.
Although manufacturers are not completely satisfiedwith the touchscreen manufacturing options available today, they do havemany options for tradeoff of cost, thickness, weight, performance, andaesthetic. They can choose from thousands of suppliers, all offering tosatisfy some portion of the touchscreen supply chain.
So, whilethe market demands development of new technologies, manufacturers muststill consider how they want their touchscreen to perform, how muchflexibility they want in their touchscreen supply chain, and they mustunderstand the technical tradeoffs that come with new technologyofferings.
Trevor Davis is a Cypress veteran who has worked within several different product groups at Cypress.
He has most recently led the World Wide Business Development organization for Cypress’s True Touch products and is now the Regional Sales Manager for the Rocky Mountain region in the USA. Before TrueTouch, Trevor served in several different Product Marketing and Business Development leadership roles for Cypress Programmable System on Chip (PSOC), USB products, Network Search Engines, CPLDs, and Software products.
Trevor received his undergraduate degree from the United States Air Force Academy and also holds his Masters in Business Administration. Trevor has worked in high technology positions for the military, nonprofit, and commercial sectors. Trevor lives in Seattle, WA and can be reached at.