A neutral earthing system protects three distinct things in an industrial installation, and confusing them is the most common reason earthing gets under-designed. First, it protects personnel by giving a low-impedance path for fault current to flow back to source fast enough to trip protective devices before a person becomes the fault path themselves. Second, it protects equipment by limiting the voltage rise across insulation during a fault, so a single earth fault does not escalate into a phase-to-phase fault or destroy the winding it started in. Third, it protects system stability by defining how the neutral behaves during a fault, whether the system rides through a single earth fault (resistance or resonant earthing) or trips immediately (solid earthing), which directly affects plant uptime and the scale of damage a fault causes before it is cleared. A neutral earthing system is not a single wire from a transformer star point to a rod in the ground; it is a designed system with a specific earthing method chosen for the plant's voltage level, fault current tolerance and continuity requirements, and the wrong method for the application either fails to protect what it should or trips the plant more often than it needs to.
"Earthing" as a term gets used loosely on most industrial sites to cover everything from a single earth rod outside a distribution board to the full neutral earthing scheme of an 11kV transformer. That looseness causes real design mistakes, because equipment earthing (bonding exposed metal parts of equipment to earth so they cannot become live) and system neutral earthing (deliberately connecting the source neutral to earth through a defined impedance) are two different protective functions that happen to share the word "earthing" and the same physical earth grid. A plant can have excellent equipment earthing, with every panel, motor frame and cable tray properly bonded, and still be exposed to serious risk if the neutral earthing method at the transformer or generator is wrong for the plant's fault current withstand and protection coordination.
This matters commercially, not just technically. A plant that experiences repeated nuisance trips on minor earth faults because it is solidly earthed when a resistance-earthed system would have ridden through the fault is losing production for a fault severity that didn't need to stop anything. A plant that is resistance earthed with a limiting resistor sized for the wrong current, or with protection relays not coordinated to the resulting fault current, is running with earth fault protection that looks present on the single line diagram but will not actually clear a fault in time. Both are earthing design failures that a periodic earth resistance test alone will not catch, because that test measures the earth electrode's resistance to true earth, not whether the neutral earthing method and its protection coordination are correct for the installation.
To see what neutral earthing protects, it helps to be specific about what an earth fault does in a system that lacks it, or has it wrongly designed.
In an unearthed system (a neutral with no deliberate earth connection at all), a single line-to-earth fault does not immediately trip anything, because there is no low-impedance return path for fault current to flow through and be detected by standard overcurrent protection. The system keeps running with one phase now at close to full line-to-line voltage relative to earth, which is a sustained overvoltage stress on the insulation of every other connected device on that system, not just the faulted one. If a second earth fault develops on a different phase before the first is found and cleared, the result is a phase-to-phase fault through earth, at full fault current, with no controlled path and no coordinated protection response, and it typically does serious damage to whichever equipment sits at the weakest point in that path. Unearthed systems are occasionally used deliberately in specific process-continuity applications (some mining and marine systems, for instance) precisely because they ride through a first fault, but that continuity comes at the cost of exactly this hidden overvoltage risk if the first fault is not found and cleared promptly, and it requires dedicated insulation monitoring to be a defensible design rather than a gap.
In a system that is earthed, but through the wrong method or impedance for its actual fault current and protection setup, a fault either produces so much current so fast that it causes arc-flash and equipment damage before protection can operate (undersized let-through coordination on a solidly earthed system), or produces so little fault current that standard protective relays cannot reliably distinguish a genuine fault from normal system noise (an over-resistance earthing scheme with relays not adjusted to the resulting low fault current). Both failure modes look, on a drawing, like the plant "has earthing." Neither actually protects what earthing is meant to protect.
The touch and step voltage a person is exposed to near a fault is a function of how much fault current flows and for how long, not just whether an earth connection exists. A correctly designed neutral earthing system, sized and coordinated with the plant's protective relaying, ensures that a fault produces enough current to be detected quickly by an earth fault relay or a residual current device, and that the resulting touch voltage during the brief period before the breaker opens stays within the safe limits the earthing design was calculated against. This is why earth fault protection settings and neutral earthing impedance are not independent decisions; the earthing method has to produce a fault current level the protection scheme can actually see and clear within the time the safety calculation assumes. An earthing system that limits fault current so heavily that protection cannot detect it in time protects equipment from destructive fault current but stops protecting personnel from the same fault, because the fault simply persists longer.
Every earth fault that is cleared quickly, at a controlled current magnitude, is a fault that did not get the chance to escalate into a larger and more destructive fault, and did not sustain an overvoltage on the rest of the system's insulation for an extended period. Resistance earthing, common in many Indian industrial plants at 11kV and 6.6kV levels, deliberately limits earth fault current to a defined value (often in the range of a few hundred amps, chosen against the specific transformer and cable thermal withstand) precisely so that a fault damages a small, contained area, typically a single cable termination or winding section, rather than propagating. This is a direct equipment protection function of the neutral earthing design, and the resistor's current rating, together with its time rating (how long it can carry that current before it must be allowed to cool), is an engineering decision tied to the specific transformer and protection scheme, not a generic catalogue selection.
This is the operational dimension of neutral earthing that gets the least attention until a plant experiences it directly. A solidly earthed system will generally trip on the first earth fault, because fault current is high and protection is designed to clear it fast, prioritising damage limitation over continuity. A resistance-earthed system, depending on the resistor value chosen, can be designed to either trip on the first fault (low-resistance earthing) or alarm and allow continued, closely monitored operation while the fault is located and repaired (high-resistance earthing, generally reserved for lower voltage systems and specific continuity-critical processes). Choosing between these is a plant-level decision about the actual cost of an unplanned trip versus the actual cost of allowing an underlying insulation fault to persist under monitoring, and it should be made deliberately at the design stage rather than defaulting to whatever earthing method the transformer happened to ship with.
| Method | Fault Current Level | Typical Application | What It Prioritises |
|---|---|---|---|
| Solid earthing | High (limited only by system impedance) | LV distribution, many industrial 415V systems | Fast, unambiguous fault clearance |
| Resistance earthing (low) | Moderate, controlled | 11kV/6.6kV industrial systems, generators | Limits fault damage, trips on first fault |
| Resistance earthing (high) | Low, controlled | Continuous-process plants at lower voltage | Ride-through of first fault with alarm, continuity |
| Resonant/Petersen coil earthing | Near-zero (tuned to cancel capacitive current) | Overhead line-heavy MV networks, some utilities | Self-extinguishing transient faults, continuity |
| Unearthed | Effectively undetectable by standard protection | Legacy or specific continuity-critical systems only | Continuity, at the cost of hidden overvoltage risk |
IS 3043, the Indian Standard Code of Practice for Earthing, is the primary reference for earthing system design in Indian industrial installations, and it covers considerably more ground than the single earth rod detail most site teams associate it with. IS 3043 sets out earth electrode design and resistance calculation methods, the design of earthing conductors and their sizing against expected fault current and clearance time (so the conductor itself survives the fault it is meant to help clear), the interconnection of equipment earthing with the system's neutral earthing, and the design principles for earth grids in substations and large industrial sites where a single rod is not sufficient. It also addresses the distinction between the functional earthing needed for the power system's neutral point and the protective earthing needed for exposed conductive parts of equipment, which is the same distinction site teams often blur when they talk about "earthing" as one undifferentiated topic.
A plant's neutral earthing method itself (solid, resistance, or resonant) is typically decided at the system design stage against the transformer's connection, the voltage level and the plant's fault current and continuity requirements, informed by the same body of practice IS 3043 codifies, while the physical earth grid, conductor sizing and electrode arrangement that the neutral earthing scheme relies on to actually dissipate fault current into the ground are designed directly to IS 3043's methods. Getting an earth resistance test certificate for the site's earth electrodes is a necessary check, but it verifies only the electrode's connection to true earth, not whether the neutral earthing method chosen upstream of that electrode is the right one for the plant's fault current and continuity needs.
Industrial sites with DG sets or other standby generation add a layer of complexity that is often missed during initial earthing design and again during later capacity additions. When a generator operates in parallel with the grid supply, or when it takes over as the sole source during a mains failure through a changeover arrangement, the question of which source's neutral is actually earthed, and whether both sources can be earthed simultaneously without creating a circulating current path, becomes a real design decision rather than an afterthought.
A common and correct arrangement on sites with automatic changeover is to earth only the source that is currently active, using a switched neutral or a four-pole changeover scheme, so that the standby generator's neutral is only connected to earth while it is actually supplying the load, and the grid neutral earthing is isolated during that period. Getting this wrong, for instance by leaving both the transformer neutral and the generator neutral solidly earthed at the same time in a three-pole changeover scheme, can create unintended parallel earth paths that confuse earth fault protection, allow fault current to circulate through unexpected routes, and in some configurations create a shock hazard that standard protection was never designed to detect. This is a detail that gets missed specifically because it sits at the boundary between the electrical contractor who installed the original transformer earthing and the generator supplier who commissioned the DG set later, with no single party reviewing the combined system.
Any site adding or upgrading standby generation capacity should treat the neutral earthing and changeover arrangement as part of that project's electrical design scope, not as a detail the generator vendor is assumed to have handled, because the generator vendor's scope is typically the DG set itself and its immediate control panel, not the site's overall earthing coordination.
For a plant reviewing its existing earthing rather than designing new, the following questions separate a genuinely protective neutral earthing system from one that merely exists on paper: