IT system and insulation monitoring (IMD) — first-fault detection

In an IT system (see the guide on earthing systems) the supply is not directly earthed, or only through a high impedance. This serves a specific purpose: on a single insulation fault between a live conductor and earth, the installation remains in service — no dangerous earth-fault current arises as it would in a TN or TT system, because there is no low-impedance return path to earth. To safely take advantage of this, a continuously operating insulation monitoring device (IMD) is required, whose requirements are laid down in NEN-EN-IEC 61557-8.

What an IMD does

An IMD injects a small measuring signal between the network and earth and continuously monitors the insulation resistance of the entire installation relative to earth. As soon as this resistance drops below a set alarm threshold, the IMD raises a signal (visual/audible alarm, sometimes with data logging), without automatically switching off the installation.

The "no disconnection on first fault" philosophy

This is the defining principle of an IT system: a first insulation fault does not lead to a dangerous touch voltage, and the installation may (and in practice must) remain in service — a critical process (for example a continuously running climate control system or a horticultural growing process) does not need to be shut down unnecessarily because of a single insulation fault. The IMD alert is the signal to locate and resolve the fault as quickly as possible, not an emergency-stop trigger.

Why a second fault does become dangerous

As soon as a second insulation fault occurs — on a different phase or in a different part of the installation, while the first fault has not yet been resolved — a fault loop is formed between the two fault locations via earth, comparable to a fault in a TN system. This combination can indeed allow a dangerous current to flow and requires immediate automatic disconnection by the regular overcurrent/RCD protection. This is precisely why a first fault must be resolved quickly: the longer a first fault persists, the greater the window in which a second fault can lead to a dangerous situation.

Note: the exact response-time requirements and insulation- resistance alarm thresholds of a specific IMD follow from the nameplate/product documentation in accordance with NEN-EN-IEC 61557-8 — these differ per manufacturer and project configuration.

Application

IT systems are used relatively rarely in regular low-voltage installations in the Netherlands (TN-C-S/TT dominate), but occur in specific situations where continuity of operation weighs more heavily than immediate disconnection on a first fault — for example in operating theatres (medical IT systems) and some industrial process installations.

Common mistakes

Ignoring an IMD alarm "because the installation is still working anyway" — precisely this delay in repair after a first fault increases the risk that a second fault causes a dangerous situation.
Applying an IT system without a functioning IMD — without continuous insulation monitoring, the installation lacks the only mechanism that makes a first fault visible, allowing it to persist unnoticed.
Assuming an IT system is "safer" than TN/TT in an absolute sense — the benefit is continuity of operation on a first fault, not a generic reduction in electrical risk; without timely fault resolution that benefit disappears.