What Is a TVS Diode? Surge Protection Explained Simply

Last updated: October 16, 2026

The short definition is easy to find and does not help much on its own: a TVS diode is a component that conducts when the voltage across it becomes too high. The useful knowledge is knowing where “too high” sits relative to your circuit, and recognising the situations where a TVS is the wrong kind of protection entirely.

The one-line definition

A transient voltage suppressor is a semiconductor device that remains effectively open at normal circuit voltage and becomes a low-resistance path when the voltage rises above its breakdown point, diverting the excess energy away from the circuit it is protecting. It returns to its off state after the event, which is what distinguishes it from a fuse: the product keeps working.

There is a second half to the definition that datasheets reveal and marketing does not. A TVS does not hold the voltage at its rated level regardless of what arrives. It offers a lower and lower resistance as the current rises, so the voltage it clamps at depends on how much current the event pushes through it. That is why the clamping voltage on a datasheet always comes with a test current next to it.

Normal days versus bad days

Think of the part as an off switch with a very specific trigger. On a normal day it is invisible: the circuit runs at its usual voltage, the TVS sits below its breakdown point and draws only leakage. On a bad day, the voltage climbs past the trigger and the switch closes, redirecting the event through a path that can survive it.

This is a different behaviour from regulation, and the distinction matters. A regulator actively holds an output at a chosen level during normal operation. A TVS does nothing during normal operation and acts only in an emergency. If you need the first behaviour, you want a regulator; if you need the second, you want a TVS. Confusing the two leads to a part that is either always conducting or never engaged.

How the voltage threshold works

Three voltage figures appear on a TVS datasheet, and reading them in the wrong order is the most common beginner mistake. The standoff voltage is the highest continuous voltage the part can sit at without conducting meaningfully; it must be above your normal rail including tolerance. The breakdown voltage is where conduction begins, specified over a range and at a test current. The clamping voltage is what your circuit actually experiences at a defined, much higher, pulse current.

Read them in that order, from the outside in. First check that the part is comfortably off during normal operation. Then check where it starts to conduct. Then check what it lets through. A part that passes the first check and fails the third is a part that will protect nothing, while appearing perfectly compliant on paper.

Surges that come from real life

The events a TVS is chosen for usually arrive from one of four directions. A person touches a connector, producing a fast static discharge. A cable picks up energy from a nearby lightning strike. An inductive load — a motor, a solenoid, a relay coil — is switched off and its stored energy has to go somewhere. Or a power source is disturbed, as when a load is disconnected from a car’s charging system and the voltage spikes before the regulator recovers.

These differ in duration by roughly six orders of magnitude, from tens of nanoseconds to tens of milliseconds, and energy scales with duration. That is why a part specified against static discharge is not automatically suitable for a load-dump event, even if its voltage rating looks generous. Matching the part to the threat means matching the waveform, not only the voltage.

Where a TVS is placed on a board

The device goes between the protected node and a low-impedance return, as close to the point where the threat enters as layout allows. In practice that means at the connector, with a short, wide path to ground. Distance is the enemy: every millimetre of trace between the connector and the protection adds inductance, and inductance converts a fast current into a voltage that appears across the very node you were protecting.

The second placement rule is about the return path. A protection device with a long, thin or shared return route will not clamp effectively, because the current has to complete a loop and the return is part of that loop. Designers who follow only the first rule and ignore the second often find that adding protection changed nothing at all.

There is a third consideration that is easy to miss because it is not about the TVS itself: whatever the device is protecting has to survive the clamped voltage, not the original spike. If the chip behind the connector is rated only slightly below the clamp level, the protection will work exactly as specified and the chip will still be stressed. The margin between the clamp and the circuit’s limit is the number that decides whether protection succeeded, and it is usually the number nobody writes down.

Three external references help when you want to check the claims on a datasheet or a supplier page. The device class is described in the transient voltage suppressors, and the consumer product that uses the same principle is covered under surge protectors. The test waveforms that make pulse ratings comparable come from the International Electrotechnical Commission, discrete device measurement methods from JEDEC, board-level assembly expectations from IPC, and industry-level reporting from the Semiconductor Industry Association. Standards and measurement background is published by NIST and ISO.

What it cannot protect against

A TVS handles short, high-voltage events. It is not a solution for sustained overvoltage, for reverse-polarity connection, for overcurrent, or for a device that is simply rated below the voltage it is being asked to work at. An event that lasts long enough to heat the part beyond its capability will destroy it, and a correctly sourced part then fails short and loads the supply — protection for the circuit, trouble for the product.

It also cannot compensate for a poor layout, as the figure above shows, and it cannot protect a signal whose integrity it has already ruined with capacitance. In those cases the part is genuinely doing its electrical job and the design still fails, which is confusing enough that it deserves saying plainly.

Two practical consequences follow from that list. The first is that a protection device is chosen against a waveform, not a voltage, so a supplier who quotes only a standoff voltage has told you almost nothing about whether the part will survive your event. The second is that the same part can be right for one node and wrong for the node next to it, because a rail with wide tolerance and a data pair with a narrow capacitance budget impose different constraints. When someone asks whether a particular TVS “works”, the honest answer is always: for which node, against which event.

TVS sizes in plain terms

FAQ

What to do next

Write down the four parameters of the node you are protecting — normal voltage, maximum tolerable voltage, the event shape you expect, and how often it might repeat — and then choose. If the four numbers are not known yet, the part selection can wait; choosing a TVS first and fitting the circuit around it is how protection ends up nominal rather than real.

This article is published by Good-Ark, a manufacturer of protection devices. It explains the device class for readers without a design background; specifications, ratings and package options belong to the current official datasheet.