What the protective conductor has to survive
The protective conductor carries current only during a fault, but it has to carry it under the worst conditions. When a live conductor touches an exposed metal part, the PE completes the loop back to the source and carries the fault current until the protective device clears it, which may take only a fraction of a second.
For that fraction of a second the conductor has to stay below its permitted final temperature, and its impedance has to be low enough for the device to operate at all. Those two requirements set its minimum cross-section, and they pull in the same direction: both favour a conductor that is large enough.
A conductor that is too small overheats, can damage its own insulation and can spoil the very disconnection it is there to support. Sizing it correctly is a safety task, not a formality, and it starts from the fault current the circuit can produce.
Two accepted methods
IEC 60364-5-54 gives two ways to reach the same answer. The table method is quick and covers ordinary cases. The adiabatic equation is more precise and is used where the table does not fit the fault energy, or where a smaller protective conductor has to be justified by calculation.
In most building installations the table is enough. The equation earns its place on circuits with long fault durations, unusual fault currents, or a protective conductor that is not made of the same material as the live conductor. Both methods answer the same question, which is the smallest section that survives the fault without harm.
Method one: the simplified table
For a protective conductor made of the same material as the live conductor, the standard gives the following minimums.
| Live conductor, S | Protective conductor |
|---|---|
| S up to 16 mm2 | S |
| Above 16 mm2 and up to 35 mm2 | 16 mm2 |
| Above 35 mm2 | S / 2 |
So a 2.5 mm2 live conductor calls for a 2.5 mm2 protective conductor, while a 50 mm2 live conductor calls for 25 mm2. These are minimums, and other conditions can still require more, so the table is a floor rather than a target.
Where the protective conductor runs separately or is made of a different material, the table is not used and the equation applies instead. The same is true when the protective conductor is not run close to the live conductors, because its impedance will be higher than the table assumes.
Method two: the adiabatic equation
The equation checks that the conductor can absorb the fault energy without exceeding its final temperature:
S = sqrt(I2 x t) / k
Here I is the fault current in amperes, t is the disconnection time in seconds, and k depends on the conductor material, its insulation and the starting and final temperatures. Typical values from the standard are shown below.
| Conductor and insulation | k |
|---|---|
| Copper, PVC insulated | 143 |
| Copper, XLPE insulated | 176 |
| Aluminium, PVC insulated | 76 |
| Aluminium, XLPE insulated | 94 |
These values apply to conductors up to 300 mm2. Larger sizes use lower figures, because heat cannot escape as easily from the centre of a thick conductor and the permitted current for a given cross-section falls.
A higher k means the insulation can take more heat, so a smaller conductor is acceptable. That is why the same fault produces a smaller required section with XLPE insulation than with PVC.
Worked example
A 400 V circuit is protected so that a fault of 3 kA is cleared in 0.2 s. The live conductor is copper with PVC insulation, so k is 143.
S = I x sqrt(t) / k = 3000 x 0.447 / 143 which is about 9.4 mm2.
The next standard size is 10 mm2. If the live conductor is 16 mm2, the table already allows a 16 mm2 protective conductor, which also covers the calculated 9.4 mm2, so 16 mm2 is safe and consistent with the table.
If the same fault were cleared in 0.4 s, the required cross-section would grow to about 13 mm2, and the table value would still cover it. The example shows how the device's operating time feeds straight into the PE decision: the slower the device, the larger the conductor it demands.
Minimum sizes that override the calculation
A calculation can suggest a small section, but mechanical strength sets a floor that no formula can go below. A conductor that is strong enough to survive installation and to stay connected through a fault is the starting point, and the thermal check is applied on top of it.
Main protective bonding conductors have their own minimum sizes, separate from the sizing of a circuit protective conductor, and they are chosen by the earthing arrangement and the supply conductor, not by the fault current of one circuit. Mixing the two rules is a common source of error.
A separate floor applies to the enclosure and the earthing terminals themselves. Even a small circuit needs a protective path that can be connected reliably, so the smallest conductor that satisfies a calculation may still be rejected because it cannot be terminated properly. The calculation sets the thermal limit, and the installation practice sets the practical one, and the larger of the two wins.
Rules that are easy to forget
A few points decide whether the design is actually safe.
- The protective conductor should run in the same cable or route as the live conductors in most circuits, so its impedance stays low.
- Minimum sizes for mechanical strength still apply, whatever the calculation says.
- Parallel paths through metalwork must never be relied on to carry fault current.
- Every joint in the protective path must be sound, because a loose connection adds impedance exactly where it hurts.
PE sizing in GorkyCAD
GorkyCAD keeps circuits, protective devices and cable data in a single electrical model. The fault current and the disconnection time behind the adiabatic check come from the same place as the cable schedule, so the protective conductor is sized against real design values rather than estimates copied into a separate table.
Because the model is shared, a change to a device or a cable updates the figures the PE sizing depends on. You can try the editor in the browser free of charge, with no card required.
Try it free in the browser — no card →
FAQ
Can the PE be smaller than the live conductor?
Yes, within the rules. For same-material conductors the table allows a PE of 16 mm2 in the middle range and half the live cross-section above 35 mm2. Up to 16 mm2 the PE matches the live conductor.
When should I use the adiabatic equation?
Use it when the table does not cover your case, when the protective conductor is a different material or route, or when you want to justify a smaller cross-section with a calculation.
What does the k factor depend on?
On the conductor material, its insulation and the starting and final temperatures the insulation can withstand. Copper and aluminium differ, and XLPE allows a higher value than PVC.
Does the disconnection time affect the PE size?
Yes. The equation contains time, so a longer disconnection time means more fault energy and a larger minimum cross-section for the same fault current.