Product
Brand
Articles
Tools
Protection components are selected backwards from the threat. A design that picks a TVS diode because it is the usual choice, without deciding what transient it must survive, generally ends up with a part that clamps too late or a part that fails on the first event.
Identify the threat first. Three different stresses are often treated as one. An ESD event is fast, high-voltage and low-energy, and is handled at the interface. A switching or inductive transient is slower and carries more energy, and is handled at the source or at the load. A surge, such as a lightning-induced or grid-coupled event, is the highest energy of the three and is handled at the entry point. The component families overlap but are not interchangeable across these roles.
TVS diodes are the fast clamp. They conduct above a breakdown voltage and hold the line at a clamping voltage for the duration of the event. The parameters that decide a part are the standoff voltage — which must sit above the highest normal operating voltage, including tolerance — the breakdown voltage, the clamping voltage at the specified peak current, and the peak pulse power for the waveform the standard uses. A TVS chosen with a standoff voltage too close to the rail will leak or conduct in normal operation.
Varistors are the higher-energy clamp, usually metal-oxide. They absorb more energy per unit volume and cost less per joule, at the price of a soft clamping characteristic, significant capacitance, and wear: every event degrades them, and a varistor that has absorbed many surges drifts and eventually fails. They also age under continuous AC stress.
ESD-rated parts and the standard behind the number. A component described as ESD-rated is quoting its performance against a defined model and level, typically the human body model and the contact-discharge levels of the IEC 61000-4-2 family. The level is part of the claim: “ESD rated” without a model and a level is not a specification. The same applies to system-level claims, which are properties of the assembled board rather than of the component alone.
Placement is part of the protection. A clamp protects what is downstream of it and nothing else. It belongs as close as possible to the point where the transient enters, with a short, low-inductance return path; a clamp placed after a long trace leaves the trace inductance in series and lets the peak voltage through. Series impedance between the connector and the protected device — a resistor or a ferrite — is what lets a small clamp do the work of a large one.
Coordination, not a single part. Robust designs use a staged approach: a high-energy device at the entry, a series element, and a fast clamp at the sensitive pin. The two devices must be chosen so that the first does not let through more than the second can absorb, and the series element must survive the current the first device diverts.
What to check before ordering. Standoff voltage against the maximum normal rail including tolerance and ripple; clamping voltage against the absolute maximum of the protected device; peak pulse power against the specified waveform; capacitance against the signal bandwidth if the line carries data; and the failure mode, because a short-circuit failure in a protection part is a different system problem than an open-circuit failure.
This article describes general protection principles and the meaning of the relevant ratings, not the specification of any particular part. Confirm standoff and clamping voltages, pulse ratings and the applicable standard levels against the datasheet and the manufacturer’s application notes for the exact part.
Specifying protection for a board?
Send the part numbers and quantities through our RFQ form, or contact us and we will confirm availability and lead time. Our catalogue lists circuit protection parts by category.
