Every time you touch the phone, the screen knows exactly where your finger is even though the surface is just a flat glass panel. The secret lies in a transparent electrode grid under the glass and in a property few people notice: the human body conducts electricity.

Most smartphones today use capacitive touch: instead of measuring the force of pressure, it measures the very small change in electrical charge that the finger causes on the electrodes.

Capacitive touch panel of a phone, tilted under light to see transparent electrical conductive lines arranged in a grid

Capacitors and fingers

According to Texas Instruments (TI) technical documentation, any two conductors placed close enough together, with an insulator in between, form a capacitor. Its ability to store electrical charge is called capacitance.

A capacitive touch panel consists of an insulating layer such as glass, coated with a transparent conductive material, usually indium tin oxide (ITO). When touched by a finger, the electrostatic field on the surface is distorted and this distortion is measured as a change in capacitance. There are two main ways of measuring:

• Self-capacitance: measures the capacitance of each electrode compared to ground. Approaching the finger causes the capacitance to increase, typically by about 1 to 10 pF.
• Mutual capacitance: measures the capacitance between two electrodes placed diagonally across each other. The human body is connected to the ground, so a finger inserted in the middle has the same effect as placing a grounded object there, causing the capacitance to decrease, usually below 1 pF.

Continuous scanning row and column grid

Phone screens using the reciprocal capacitive type mainly arrange electrodes into a grid: horizontal lines are the emitters (TX), vertical lines are the receivers (RX), or vice versa. Each intersection between a row and a column is a small capacitor.

The controller applies high-frequency voltage pulses to each row in turn. This pulse is transmitted to every column that crosses it, and the controller measures the signal received at each column. Wherever the signal is significantly weaker, there is a finger.

Illustrative diagram: the controller in turn pulses on each TX row, measures the signal in the RX columns and finds the node with reduced capacitance

Because each intersection point is measured separately, the controller determines multiple touch points at the same time, the basis of multi-touch operations such as using two fingers to zoom in on an image. Wikipedia gives an example of a 16 x 14 grid with 224 intersections, or 224 capacitors; The more intersections, the higher the resolution.

Fingers often cover several buttons at the same time. The controller uses an algorithm to calculate the center of gravity, taking the signal of the strongest node and neighboring nodes to interpolate the position. According to Infineon's design documents, a touch slider with only 5 electrode segments can still distinguish at least 100 finger positions.

Advertisement

Why is it difficult for self-capacitance to recognize multiple fingers

Self-capacitance can also be arranged in rows and columns, but each row and each column are measured independently. If you touch two fingers at two diagonal corners, the controller only knows that two rows and two columns are "light", not distinguishable from the other two diagonal corners. TI calls this ghosting phenomenon. In return, self-capacitance is more sensitive, so it is often used to detect fingers hovering near the screen.

From accelerator control room to pocket

The first finger-controlled touch screen was developed by Eric A. Johnson at the Royal Radar Establishment (UK) and described in a 1965 paper. At CERN, engineer Bent Stumpe proposed a capacitive touch screen in a handwritten note dated March 11, 1972 to control the SPS accelerator. The capacitors are copper lines etched into glass, 80 µm wide, thin enough to be invisible.

Two early capacitive touch screens developed by CERN: on the right is the self-capacitive screen of 1972, on the left is the x-y grid prototype of 1977

When SPS started up in 1976, the control room was fully equipped with touch screens, which according to the CERN Courier was probably the first application of capacitive touch screens in the world. In the late 1970s, Stumpe continued to make a version of the x-y grid using reciprocal capacitance.

Familiar limits

Advertisement

• Plastic gloves and pen: insulating material blocks the influence of fingers on the electric field, so a capacitive touch pen or gloves with conductive fibers at the fingertips are needed.
• Poor quality charger: sources with noisy voltages may temporarily reduce touch accuracy and sensitivity.
• Thinner: in-cell technology, like on Samsung's Super AMOLED screen, puts the touch layer into the display panel.

Atmel mXT1386 ​​touch control chip mounted on the touch panel of the Motorola Xoom tablet, which processes the signal from the electrode grid

According to Wikipedia, the controller is usually a specialized CMOS chip, which converts signals to a digital signal processor to calculate the touch position. Each seemingly simple touch is actually the result of hundreds of capacitance measurements that differ by less than a picofarad.

Source:

Wikipedia

,

Texas Instruments

,

CERN Courier

,

Infineon

,

Explain that Stuff