DC switching devices — why an AC-rated breaker or switch cannot simply be reused on a DC circuit

The guide on DC arc-flash and arc suppression covers the incident energy released by a sustained DC arc, and the guide on DC arc-fault detection in PV strings covers detecting an unwanted series arc within a string. This article covers a more basic, prior question: why the switching device itself — a circuit-breaker, a load-break switch, a disconnector — needs a fundamentally different internal design on a DC circuit than on an AC circuit of the same voltage and current, even when it is used purely to make and break a circuit under normal, healthy conditions.

The core difference: the AC current zero-crossing

An alternating current, at 50 Hz, passes through zero twice every cycle — once every 10 ms. When a pair of contacts separates and an arc forms between them, that arc self-extinguishes at (or very near) each natural current zero, because the current available to sustain ionisation of the arc path momentarily disappears. An AC switching device only has to prevent the arc from re-igniting immediately after that zero-crossing — a comparatively modest task, achieved with a contact gap and arc chute sized for that purpose.

A direct current has no natural zero-crossing. Once an arc is struck between separating DC contacts, it keeps burning — fed continuously by the source — for as long as the arc voltage stays below the source voltage. The device must therefore actively force the arc to extinguish, rather than simply preventing its re-ignition after a zero that never arrives.

How a DC-rated device forces arc extinction

To interrupt a DC arc, a switching device has to stretch, cool and split the arc until its own voltage (which rises as the arc lengthens and is divided into multiple shorter arcs) exceeds the source voltage — at that point the arc can no longer sustain itself and is driven out. In practice this is achieved through a combination of:

• A larger contact gap and contact travel than an equivalent AC device, so the arc is stretched further before extinction is even attempted.
• A purpose-built arc chute with more (and often more closely spaced) splitter plates than an AC arc chute of comparable current rating, to divide the single arc into several shorter series arcs, each adding its own arc voltage.
• Frequently a magnetic blow-out coil, which uses the magnetic field generated by the fault/load current itself to actively drive the arc into the arc chute, rather than relying only on convection and the natural elongation of the arc as the contacts separate.

Why poles in series matter for higher DC voltages

Because a single pole's contact gap and arc chute can only force extinction up to a certain DC voltage, many DC-rated miniature circuit breakers and switch-disconnectors — for example on a PV DC string circuit with an open-circuit voltage well above 600 V — are applied with two (or more) poles connected in series, in one or both polarities. Each pole in the series chain contributes its own arc voltage when it opens, so the combined arc voltage of the series string can exceed the source voltage even though no single pole could do so alone. A device's DC voltage rating on the nameplate already reflects the pole configuration it was tested in (for example "500 V DC with 2 poles in series") — that rating does not automatically apply if the same device is later wired with a different number of poles in series.

IEC 60947-3 DC utilisation categories

For switches, disconnectors and switch-disconnectors, IEC 60947-3 defines DC utilisation categories that parallel the familiar AC ones:

A device rated for DC-20 (isolation only) may not be used to make or break a live DC-22 or DC-23 load — the no-load rating says nothing about its ability to interrupt an arc under load.

Why an AC-rated device is not simply "safe to reuse" on DC

A breaker or switch that carries only an AC rating on its nameplate has not been tested for DC arc interruption at all. Applying it on a DC circuit anyway — a mistake that occurs in practice when retrofitting a battery storage system or a PV DC combiner box with whatever breaker happens to fit the enclosure — risks a sustained arc that the device cannot extinguish: the contacts weld together, the enclosure overheats, or the arc persists until an upstream device (if any) finally interrupts it, with far higher incident energy than the same fault would have produced on an AC circuit of similar voltage and current (see the DC arc-flash guide for that consequence in more detail).

Practical relevance

When specifying or inspecting switching devices for a DC installation — a PV DC side, a battery energy storage system, a DC microgrid — the nameplate must show an explicit DC rating (voltage, current, and utilisation category), not only an AC rating, and the number of poles actually wired in series must match the pole configuration for which that DC voltage rating was tested.

Common mistakes

1. Applying a breaker or switch with only an AC rating on a DC circuit, assuming that "a contact is a contact" — without a tested DC rating, the device's ability to force extinction of a sustained DC arc is unknown and often absent.
2. Wiring poles in series without checking that the resulting configuration matches the manufacturer's tested DC voltage rating — a device's DC voltage figure is only valid for the specific pole configuration in which it was verified.
3. Confusing the DC-20 (no-load, isolation-only) category with a load-break rating — a device suitable only for isolating a de-energised circuit must not be used to interrupt a live DC-22 or DC-23 load.
4. Assuming a magnetic blow-out coil is optional decoration rather than a functional part of the arc-extinction mechanism on many DC-rated devices.

Related

• DC arc-flash and arc suppression
• DC arc-fault detection in PV strings
• Moulded-case circuit breaker (MCCB) versus miniature circuit breaker (MCB)
• PV fire-fighter's switch