Protecting against ESD

An effective approach for protecting any electronic system against ESD is to minimize mechanically the pathways by which high voltages enter the system from the outside environment. This can be especially difficult if the user needs to touch the device, or if there is a void or other openings in the packaging.

Mechanical switches and control potentiometers are classic system entry points for ESD stress. Changing from mechanical controls to capacitive touch controls eliminates the voids for these traditional ESD entry paths.

The elements listed below are used to protect against ESD stress:

• dielectric overlays (see Dielectric overlays)
• spark gaps (see Spark gaps)
• ground rings (see Ground rings)
• additional resistance (see Additional resistance)
• additional diodes (see Additional diodes)
• ESD protection devices (see ESD protection devices)
• firmware (see Firmware)

In the touch sensing application domain, a protective layer of dielectric material (any insulating material that can intrinsically withstand high voltages without breaking down) can be placed between the ESD source and the touch sensing application. For example, one layer of 5 mil Kapton® tape withstands 18 kV. Other dielectric overlay materials are listed in the table below, together with their dielectric strengths.

The use of the dielectric overlay is effective and is almost mandatory for many applications. However, the overlay does not surroundcompletely the device, and an ESD event can bypass it. If the user can avoid accessing the application through the front panel (example, by accessing it from the back or elsewhere) or if the ESD event can bypass the front panel, one of the other methods protecting the device against ESD must be considered.

Physical techniques, such as the addition of spark gaps, can give supplementary protection to the input/output lines of a circuit board susceptible to extraneous voltage such as ESD. For example, printed circuit board (PCB) spark gaps can be used to route ESD to earth in products using capacitive sensing electrodes (see the figure below).

A spark gap consists in an arrangement of two conducting electrodes separated by a gap usually filled with a gas (such as air), designed to allow an electric spark to pass between the conductors. When the voltage difference between the conductors exceeds the breakdown voltage of the gap, a spark forms, ionizing the gas, and drastically reducing its electrical resistance.

The spark gap shown in the above figure is 8 mm, a common PCB tolerance. The approximate breakdown is given by the equation V = (3000 x p x d) + 1350, where p is the pressure in atmospheres, and d is the distance in millimeters.

This spark gap can be expected to have a peak voltage `between 2000 and 2500 V.

To protect against ESD stress on the touch sensing surface, a low impedance path to ground must exist through the device. The touch sensor can be protected using a ground (also called a guard) ring, which is a ring around all the system electrodes (see Figure 4 Ground ring), placed along the border. The ground ring can be a simple metal foil. It is necessary to ensure a firm connection between the ground ring and the device system ground.

These three items avoid an ESD stress reaching the electrodes (and therefore the MCU) because they form a sufficient shield. For example, an ESD stress goes directly to ground if it occurs in the ground ring.

As recommended in AN4312 Guidelines for designing touch sensing applications with surface sensors, a hatched ground plane around the touchkey or rotary or linear sensors (other types of electrode with different shapes) can redirect the ESD stress away from the electrodes and touch sensing device.

The most common method of external ESD protection is to add a small serial resistor in-line between the ESD energy source and the touch sensing device pin to be protected. A resistor as small as 50 Ω can double the ESD immunity of a CMOS device. A higher level of protection is somewhat proportional to increased serial resistance so, higher immunity is possible.

This method works for the following reasons:

• The serial resistor works with the parasitic pin capacitance (typically 5 to 10 pF) of the device to create a single-pole low-pass filter with a cutoff frequency below 1 GHz. This causes the serial resistor to attenuate most of the high-frequency energy of an ESD event (as much as 90 % of the rising-edge power in an HBM discharge).
• When the protection circuits of the device are operating normally, their impedance is very low (in the order of tens of ohms or less). This low impedance works with the serial resistor to create a voltage divider, so that the high voltage from an ESD stress can only bias the built-in protection circuits of the device with a portion of the total ESD voltage. This attenuation is in addition to rising-edge filtering. The sum of these effects from a simple external serial resistor dramatically improves ESD performance in a demanding application.
• Usually for touch sensing, a 10 kΩ sampling resistor R_S is used to improve conducted noise immunity (IEC61000-4-6).

Input/output lines that are susceptible to ESD stress are sometimes protected by adding external diodes that shunt the high energy of the ESD stress before it can reach the device input pin. These diodes may either pass the current to the power-supply rails or may internally dissipate the unwanted power. External diodes are similar to the diodes built into a device (internal diodes) for protection but they are designed differently. External diodes have the following significant advantages:

• They can switch faster and at a lower excursion voltage than the internal diodes of the device.
• They can have much better connections to the supply rails and can carry more power.

The effects of external diodes on circuit operation are different from internal diodes, because the connections used internally cannot be achieved with external devices.

The following types of protection diode are typically used against ESD stress:

• Transient voltage suppression (TVS) diodes can be placed between an input signal and ground. In this configuration, the diode protects the CMOS input by reverse conduction whenever its voltage rises above the specified diode breakdown voltage. Negative ESD excursions are shunted to ground through normal diode action.
• Diode pairs (such as Schottky diodes, with lower forward voltage drop) are placed between the input line and the power and ground rails. These devices protect the CMOS input by normal diode conduction whenever the input line voltage moves outside the range of the power supply rails.

For touch sensing, an additional Schottky diode (like ST BAS70), can be used to improve conducted noise immunity (IEC61000-4-6). ESD, in this case, is not the main concern.

Diodes placed on capacitive sensed lines present the same problems to capacitive sensing circuits as they do with any analog circuit input: they can be highly capacitive (over 100 pF) and leaky. Some Schottky pairs leak over 20 µA; some avalanche diodes leak over 1 mA when operated near their reverse-standoff voltage (generating significant noise voltage as well).

Although these given numbers are for the least suitable devices, the most commonly-used schottky and TVS diodes have parasitic parameters that make them unacceptable for use in capacitive sensing applications. If the diode circuit can be designed to add only a very small amount of additional capacitance, capacitance sensing solutions can be adjusted to match.

Compensation mechanisms are usually built into the touch sensing device for adaptation to the naturally-occurring changes in capacitance that result from environmental changes. However, leakage and bulk capacitance can create problems for any sort of capacitive sensing method, some more than others.

External diodes with high reverse leakage make the test capacitance look larger because their leakage drains test current from the circuit. This disappearing test current (which should fill the capacitance under test) has no dV/dt effect on the test load.

As diode leakage currents approach the level of the test current, the apparent load capacitance approaches infinity. Also, the amount of current required to detect a 0.1 pF change in capacitance is less than 20 pA, many orders of magnitude less than the leakage current for some protection diodes. For this reason, where external diodes must be used, it is essential to specify devices with extremely low reverse leakage.

The ST ESD protection diodes are designed to resist multiple ESD stresses. Their low capacitance (1 pF) and low leakage (less than 100 nA) reduce the problems encountered with Schottky protection diodes. The bidirectional protection ESDAXLC6 diode, with even lower capacitance (0.5 pF) can be used instead of the ESDAULC6 diode to prevent the occurrence of negative and positive pulses.