Temporary overvoltage from an earth fault in the high-voltage network (§442) — why the substation voltage rise reaches the low-voltage installation

The SPD overvoltage protection guide covers transient overvoltage — brief spikes from lightning strikes or switching operations, handled by an SPD under §534. §442 covers a fundamentally different kind of overvoltage: a temporary, power-frequency voltage rise caused by an earth fault in the (medium) high-voltage network connected to the low-voltage installation via the same transformer substation.

How a high-voltage network fault reaches the low-voltage installation

During an earth fault in the medium-voltage network (for example a breakdown of an MV cable to earth), an earth fault current flows through the earthing arrangement of the transformer substation. Via the earthing resistance of that arrangement, this fault current causes a temporary rise in potential — denoted Uf — at the substation's own earthing.

When the neutral (N/PEN) of the low-voltage network is connected to that same substation earthing arrangement — as is common in a TN system — this temporary rise in potential propagates directly into the low-voltage installation, even though the actual fault lies entirely on the medium-voltage side.

What the standard regulates: magnitude and duration, not one hard limit

§442 states that the magnitude and duration of the power-frequency voltage across the insulation of low-voltage equipment, caused by an earth fault in the high-voltage network, must not exceed the values in table 44A of the standard. The underlying principle: the higher the resulting voltage Uf, the shorter it may persist before the fault in the high-voltage network must be cleared — a brief, high spike can be acceptable where a prolonged elevated voltage would not be.

Note: this article gives the underlying principle (a magnitude-duration relationship, tied to Uf and the disconnection time of the high-voltage protection). The exact voltage values and permitted disconnection times from table 44A itself must be looked up in the full standard for a concrete assessment, not reconstructed from memory.

Why this matters to the installer, not only the network operator

At first glance this looks like a topic entirely owned by the network operator — the earth fault itself occurs in the medium-voltage network, far from the installation itself. Yet it is relevant to anyone designing or assessing a low-voltage installation:

The connection between the low-voltage network's neutral and the transformer substation's earthing arrangement (characteristic of TN) is precisely the mechanism through which a temporary overvoltage from the high-voltage side can propagate into the installation.
This is a separate, additional reason to correctly assess the earthing arrangement and the connection to the network, distinct from the usual considerations around earth fault current and fault voltage within the installation itself.

Practical relevance

For most day-to-day installation work this is not a topic that is actively calculated or measured — the assessment of Uf and the disconnection time of the high-voltage protection is in practice a matter for the network operator and the substation designer. For an installer, it is mainly relevant as background knowledge: it explains why the network's earthing arrangement and the own installation are not entirely independent of each other, even though the own low-voltage installation at first glance looks like a closed, independent system.

Common mistakes

Assuming a fault in the medium-voltage network can never affect the own low-voltage installation — in a TN system with a shared earthing arrangement, that mechanism is very much present.
Confusing §442 with §534 (SPD/transient overvoltage) — §442 concerns a temporary, power-frequency voltage rise via the earthing arrangement during a high-voltage fault, not a brief spike from lightning or switching.
Assuming a single, universal voltage value always applies — the standard explicitly ties the permitted voltage to the disconnection time of the high-voltage protection (table 44A), not to a fixed, time-independent limit.