Commissioning an Isolated Power Panel: The Checklist

An isolated power panel can look finished the day the electrician walks off site — lamps lit, LIM display reading zero, breakers labelled — and still not be commissioned in any sense NFPA 99 would recognise. Commissioning is a specific set of measured tests with pass values attached to them, not a visual check that the equipment is installed and powered. The following is the sequence we work through, with the actual thresholds the standard sets for each step.

1. Insulation Resistance, Before the LIM Is Even Connected

The first test happens before the line isolation monitor and any suppression equipment are connected at all. Each isolated conductor is tested to ground, and the resistive impedance has to measure at least 1 megohm. This establishes a clean baseline for the wiring itself, separate from anything the LIM or connected equipment will later contribute to the system's total leakage.

Skipping this step and only testing after everything is connected hides problems rather than catching them. If the finished system later reads low, there is no baseline to tell you whether the wiring was ever actually sound, which turns a five-minute isolation question into a much longer fault-finding exercise.

2. System Impedance to Ground, as Installed

Once suppression equipment and the LIM are connected, the impedance — capacitive and resistive together — from either isolated conductor to ground has to exceed 200,000 ohms for the installed system. This is a different, more conservative figure than the 1 megohm bare-wiring test, because it reflects the system as it will actually operate, including the small amount of leakage every piece of connected equipment legitimately contributes.

This is also the number that degrades over time as circuits get added to a theatre, which is why it is worth recording precisely at commissioning rather than just noting a pass. A baseline reading lets anyone checking the system years later tell whether a falling number reflects normal accumulated load or an actual developing fault.

3. Line Isolation Monitor Calibration — the Two Numbers That Matter

The LIM itself gets tested by successively grounding each isolated conductor through a resistor sized at 200 × V ohms, where V is the measured line voltage. This simulates a real fault of known severity and confirms the monitor responds at the correct point, rather than relying on the device's factory calibration being correct out of the box.

NFPA 99 sets two thresholds, and both have to be verified, not just the alarm point. The monitor has to alarm when total hazard current reaches 5.0 mA. It also has to not alarm below 3.7 mA — a monitor that goes off at 2 mA has not passed this test even though it did alarm, because an over-sensitive LIM trains staff to ignore the red lamp as background noise, which defeats the point of having one.

We cover what these two numbers mean for day-to-day operation, as opposed to commissioning, in more detail in our piece on NFPA 99 and line isolation monitors. For commissioning purposes, both numbers need to be in the test record with the actual measured values, not just a pass/fail checkbox.

4. The LIM's Own Contribution to Leakage

The monitor adds its own small leakage to the system by virtue of measuring it, and NFPA 99 caps that contribution: when any point of the isolated system is grounded, the maximum current flowing through the LIM itself has to stay at or below 1 mA. A LIM that contributes more than this is eating into the margin the rest of the system has before it reaches the 5.0 mA alarm point, for no operational reason.

This is one of the figures that is genuinely difficult to check without the right test equipment, and it is also one of the figures most often skipped in a rushed commissioning, because the panel will still pass a basic functional test without it. It is worth specifically asking whoever commissions a panel whether this was measured, not assumed.

5. Equipotential Grounding

Separately from the isolated power circuit itself, the equipotential grounding system — the network that keeps every conductive surface in the room at the same potential — has its own impedance limit: 0.1 ohm for a standard system, relaxed to 0.2 ohm where isolated-ground receptacles are present, with a voltage limit of 20 mV between any two points.

This test gets missed more often than the LIM calibration, partly because it is a separate measurement on a separate system, and partly because a room can look and function normally with a non-compliant equipotential ground right up until the specific fault condition it exists to prevent actually occurs. It is not optional, and it is not covered by testing the LIM.

6. Functional Test, Last — Not First

Pressing the LIM's test button and confirming the lamp and buzzer activate is a legitimate part of commissioning, but only as the final confirmation after the measured tests above, not as a substitute for them. The test button places an internal fault well above the threshold being verified; it proves the alarm circuit works, not that the monitor is calibrated correctly. A LIM that has drifted to alarm at 8 mA instead of 5.0 mA still passes the button test perfectly.

A commissioning record that only documents a button-test pass, with no measured insulation resistance, no recorded system impedance, no documented LIM calibration against the 5.0 mA and 3.7 mA thresholds, and no equipotential grounding figures, is not evidence the panel meets NFPA 99. It is evidence the lamp works.

Keeping the Record

Every figure above should end up in a written commissioning record with the actual measured values, not a checklist of ticks, and that record is what gets referenced at every subsequent annual test to track whether numbers are holding steady or drifting. A facility that can produce this record when an inspector, accreditation reviewer or insurer asks for it is in a materially different position than one that can only say the panel was tested and passed.