An insulation monitoring device (IMD) works by continuously measuring the insulation resistance between the live conductors of an ungrounded (IT) power system and earth, then comparing it with a set alarm value. When insulation falls too low, the IMD raises an alarm before a second fault can cause a short circuit.

This guide is the starting point for our insulation monitoring series. It explains what an IMD measures, the step-by-step measuring sequence, the three main measuring methods (balanced bridge, unbalanced bridge and signal injection), how to set the alarm value, and where IMDs are used. Later guides go deeper into DC ground fault detection methods, EV chargers and battery energy storage.

Why do IT systems need an insulation monitoring device?

An IT system is a power system that is isolated from earth, or connected to it only through a very high impedance. Many DC control supplies, battery systems, EV chargers, hospital circuits and industrial process supplies are built this way.

The benefit is continuity. If one conductor touches earth (a first fault), only a very small current flows, so the system can keep running and people are not exposed to a large fault current.

The problem is that the first fault is invisible. Nothing trips and nothing looks wrong. If a second fault later appears on another conductor, the two faults form a short-circuit path through earth. The result can be fire, equipment damage, a shock hazard or an unplanned outage.

The IMD removes the blind spot. It detects the first fault so you can repair it while the system is still safe.

How does an insulation monitoring device work, step by step?

Designs differ, but most IMDs follow the same sequence:

1. Connect: The IMD is connected between the live conductors of the system and protective earth (PE).
2. Apply a measuring signal: It either uses a resistor network across the system voltage or injects its own low-level measuring voltage.
3. Measure the response: It reads the resulting voltage or current between the system and earth.
4. Calculate: From that reading it works out the insulation resistance. In DC systems, better devices calculate the positive pole and the negative pole separately.
5. Compare: The result is compared with the response value you set.
6. Alarm and communicate: If the resistance is too low, the IMD switches its relay and shows the value. Many models also report it over RS485 or CAN.

The main difference between IMDs is the measuring method in step 2.

What are the main IMD measuring methods?

Three methods cover most applications. The table compares them.

How does the balanced bridge method work?

Two equal resistors connect the positive pole and the negative pole to earth. In a healthy system the voltage at the midpoint stays balanced. When one pole loses insulation, the balance shifts. The IMD measures the shift and converts it into an insulation resistance value.

Its limit is the symmetrical fault. If both poles degrade by the same amount, the bridge stays balanced and the fault can go unnoticed.

How does the unbalanced bridge method work?

The IMD switches a known resistor between the positive pole and earth, then between the negative pole and earth, and records the voltage change in each state. Two different measurements allow it to solve for the insulation resistance of each pole separately.

That is why the unbalanced bridge also detects symmetrical faults that a balanced bridge misses.

How does signal injection work?

The IMD applies a low-frequency measuring voltage between the system and earth and measures the current that returns through the insulation. Because it uses its own signal, the method works on AC, DC and mixed systems.

Signal injection is also the basis of many fault-location systems, which use current sensors to follow the signal to the faulty branch. Details are in our guide to DC ground fault detection methods.

What is Y-capacitance and why does it matter to an IMD?

Y-capacitance (leakage capacitance) is the natural capacitance between live conductors and earth. It comes from long cables, EMC filters and Y-capacitors in power electronics.

It matters because large capacitance slows down the measurement and can disturb some measuring methods. Big installations, such as a battery system with many cables, therefore need an IMD that is designed for high leakage capacitance. On high-voltage systems a coupling device may be needed to connect the IMD safely.

What is the response value and how do you set it?

The response value is the resistance below which the IMD alarms. It is set in kΩ, but many standards express the requirement in ohms per volt (Ω/V) of system voltage, so that the limit scales with voltage.

Two reference values appear often: 100 Ω/V for DC circuits and 500 Ω/V for AC circuits (for example in ISO 6469-3 for electric vehicles). The kΩ value is the Ω/V figure multiplied by the system voltage.

Worked example: on a 1000 Vdc bus, 100 Ω/V gives 100 kΩ and 500 Ω/V gives 500 kΩ. A common setup is a pre-warning at the higher value and a main alarm at the lower value. The pre-warning tells maintenance to plan an inspection; the main alarm tells operators the safety margin is nearly used up.

IEC 61557-8 defines how the IMD must perform, but the response value itself comes from the application standard, your project specification and the equipment manufacturer. Always confirm the value that applies to your system.

How is an IMD different from an insulation resistance tester or an RCD?

They complement each other. The periodic megger test verifies the installation, and the IMD protects it between tests. Disconnect the IMD before high-voltage megger testing, because the test voltage can damage it.

Where are insulation monitoring devices used?

IMDs are used wherever an ungrounded system must keep running safely:

• DC EV charging stations, where the charger must confirm insulation before and during charging. See insulation monitoring for EV chargers.
• Battery energy storage systems, where high-voltage DC buses in racks and containers need continuous monitoring. See insulation monitoring for battery energy storage.
• Industrial and substation DC systems, where IMDs catch ground faults on control and battery supplies.
• Solar plants, hospitals, marine and rail systems, which all use ungrounded supplies where safety and uptime matter.

Which measuring methods does Gongyuan use?

Gongyuan (Suzhou Gongyuan Automation Technology Co., Ltd.) offers IMDs for DC and mixed systems. The GYIM-DAR DC insulation monitor supports both balanced and unbalanced bridge modes, so the engineer can choose the mode that fits the system. A balanced bridge is simple when faults are expected on one pole. The unbalanced bridge adds detection of symmetrical insulation loss.

For systems that combine AC and DC, the GYID AC/DC insulation monitoring device covers both.

What are common mistakes when applying an IMD?

• Two active IMDs on one connected network: They disturb each other’s measurement. Only one active monitor should operate on each galvanically connected IT system.
• Poor PE connection: A loose or high-resistance earth connection gives wrong readings.
• Ignoring Y-capacitance: Choose the device for the real system capacitance, not just the voltage.
• Guessing the response value: Take it from the standard and the project specification.
• Expecting an IMD to find the fault: A basic IMD reports that a fault exists, not where. Location needs branch monitoring.
• Skipping the alarm test at commissioning: Simulate a low-resistance fault and confirm the relay, display and control-system alarm all respond.

Conclusion

An insulation monitoring device measures the resistance between live conductors and earth, compares it with a response value and alarms when insulation degrades. The balanced bridge, unbalanced bridge and signal injection methods each suit different systems, and the right choice depends on system type, voltage and leakage capacitance.

For a broader overview, see the insulation monitoring device page on Wikipedia.