TVS Diode Selection and Placement for ESD Protection
Every board that connects to the outside world needs a defined answer to the question of what happens when a discharge arrives. A transient voltage suppression diode is the most common answer, because it responds in fractions of a nanosecond, absorbs a large surge current, and is available in a wide range of voltages. Using one well is mostly a matter of three decisions: which part, where it goes, and what else the circuit has to do to make it effective.
What the Device Does
A silicon transient voltage suppression diode is designed to remain effectively open at the circuit”s working voltage and to become a low-impedance path when the voltage across it exceeds a threshold. Its response time is in the sub-nanosecond range and its surge current capability is high, which is why it is used to protect equipment against electrostatic discharge, the transient produced when an inductive load is switched, and the overvoltage induced by a nearby lightning strike. Single devices protect in one direction and pairs connected back to back protect in both, and the choice follows the polarity of the threat and the nature of the signal.
The Three Parameters That Decide the Part
The first parameter is the breakdown voltage, and it has to sit above the highest voltage the circuit will legitimately see, including tolerance and temperature variation. A working rule is to choose a device whose breakdown voltage is about ten percent above the normal operating voltage. Choosing one that is too close means the leakage current through the diode, which rises sharply as the voltage approaches breakdown, may disturb the circuit; it also means that a rise in temperature, which lowers the breakdown voltage, can bring it into the normal operating range. The second parameter is leakage current, which matters most for high-impedance or low-power circuits. The third is junction capacitance, which is the parameter that decides whether the device can be used on a fast signal at all. Alongside these, the clamping voltage that appears across the device during a transient should be compared with the absolute maximum rating of the pin being protected, since a clamp that limits correctly for one device may still exceed the rating of a more sensitive one.
Junction Capacitance and High-Speed Lines
Capacitance across a signal line forms a low-pass filter with the impedance of the line, and on a high-speed interface that filter can close the eye. A single diode with a large junction area has a large capacitance, so the standard solution is to connect the diode in series with a fast recovery diode in a back-to-back arrangement. The fast recovery diode has a small junction capacitance, so the capacitance of the combination is small enough for a high-frequency line while the transient is still absorbed by the pair. The arrangement is common on data and I/O lines, and its placement matters as much as its value: a protection device connected by a long trace has already allowed the transient to travel before it acts. The rule is that the clamp belongs closer to the connector than to the integrated circuit it protects, so the disturbance is attenuated before it reaches the sensitive pin. The same principle that governs the board edge in general, described in this article on ESD and PCB edge traces, applies to the position of the diode.
Series Operation and Power Handling
When no single device has a suitable voltage rating, several may be connected in series. The current the string can carry is set by the weakest device in it, and the peak power the string absorbs is the product of that current and the sum of the voltages across the string, so the rating of the weakest part governs the whole assembly. For repeated transients, the average power the device must dissipate in steady state has to be checked against its rating, and the rating has to be derated as the ambient temperature rises. The absorbed power also depends on the duration of the pulse: a device rated for a short pulse cannot absorb the same energy from a longer one, so where the pulses are wide or poorly defined, a derating factor should be applied rather than assuming the peak figure. Where the load is small, a series resistor can be added deliberately to limit the current that reaches the clamp, which allows a smaller device to be used without affecting normal operation.
Other Devices in the Same Family
A solid state discharge tube works on a different principle from a clamping diode, but occupies a similar place in the design. Its response time is slower, in the range of ten to twenty nanoseconds, but its current capability is larger and its trigger voltage is stable over life. It behaves like a switch rather than a clamp: below the trigger voltage it is effectively open, and above it the characteristic turns over into a low-resistance region where the energy is diverted. When the current falls below the holding value, the device returns to its high-resistance state and is ready for the next event. Its distinguishing feature is the failure mode, which is a short circuit between the electrodes, and that is a requirement in applications where an open-circuit failure would leave the equipment unprotected without anyone noticing. The trade-off is a narrower choice of voltages, which makes it suitable for line protection in network and communication equipment and for protection at the assembly level rather than at the individual signal.
Package and Placement
The package should be chosen from where the device sits. An axial leaded part suits a supply rail, where the current is larger and the layout is less dense. Surface mount and dual in-line packages suit logic circuits, input and output buses and data buses on a printed board, where the space is tight and the connection has to be short. Whichever is used, the length of the connection to the protected line and the distance from the connector are the two layout variables that most affect the result, and both should be minimised. The ground connection of the clamp deserves the same attention: it should return to a plane or to a wide trace immediately beside the device rather than through a shared, narrow path, because the impedance of that connection is in series with the transient the device is trying to divert. Combining a clamp with a carefully laid out board edge and a wide, unbroken ground on the connector side gives a level of ESD protection that no single device can provide on its own, and the general principles are described in this article on EMI suppression design principles. The pad geometry for the fine-pitch packages used on data lines follows the usual land pattern rules, described in this article on PCB pad design standards.
FAQ
How far above the working voltage should the breakdown voltage be? About ten percent, which keeps leakage current low at the normal operating voltage and leaves room for the breakdown voltage to fall as the temperature rises.
Why pair a TVS diode with a fast recovery diode? To reduce the capacitance presented to a high-speed line. The fast recovery diode has a small junction capacitance, so the combination is suitable where a single device would be too slow.
What is the advantage of a solid state discharge tube? A defined short-circuit failure mode and a stable trigger voltage, which makes it useful where an unnoticed open-circuit failure would leave equipment unprotected.
