Electronics Guide

Earthing Versus Bonding

Earthing connects a part of an electrical system, or the exposed metalwork of equipment, to the earth through an earth electrode. Its purpose is to fix the potential of that part relative to the surrounding ground and to provide a return path for fault and leakage currents. Bonding, by contrast, connects conductive parts together to equalize their potential. Equipotential bonding ensures that, even during a fault, the voltage between two touchable surfaces remains low enough to avoid a dangerous shock, regardless of the absolute potential relative to remote earth.

The distinction matters because earth is not a perfect conductor. The resistance of an earth electrode is rarely negligible, so during a large fault the local ground can rise well above remote earth potential. Bonding addresses this by keeping nearby objects at a common potential, so a person standing in the affected area does not bridge a hazardous voltage difference. Earthing and bonding therefore complement each other: earthing references the system, while bonding protects the person.

Fault Loop Impedance and Disconnection

Protective earthing depends on a low-impedance fault loop. When a live conductor contacts an earthed enclosure, current flows through the protective conductor and back to the source. If this loop impedance is low, the resulting fault current is large enough to trip a circuit breaker or blow a fuse within the time required by the applicable standard. Excessive loop impedance allows only a modest fault current, which may not operate the overcurrent device quickly, leaving the enclosure energized at a dangerous voltage.

Disconnection-time requirements link loop impedance to safety. IEC 60364-4-41 tabulates maximum disconnection times by earthing arrangement and nominal voltage. In a TN system at 230 volts to earth, a final circuit rated up to sixty-three amperes that supplies socket outlets must be cleared within 0.4 second, while distribution circuits are allowed up to five seconds. The corresponding limits for a TT system, whose fault loop runs partly through soil, are 0.2 second and one second. The designer verifies that the measured or calculated loop impedance, combined with the time-current characteristic of the protective device, satisfies these limits. Where loop impedance cannot be made low enough, residual-current devices provide the necessary protection by responding to leakage current rather than relying on a high fault current.

Touch Voltage and Step Voltage

Safety is ultimately about the voltage that appears across a human body and the current that results. Touch voltage is the potential difference between an energized part and a point a person can reach, typically the ground on which the person stands. Step voltage is the difference between two points on the ground separated by a normal stride, which becomes significant near grounding electrodes carrying large fault or lightning currents. Equipotential bonding and properly graded earthing reduce both touch and step voltages to tolerable levels.

The tolerable level depends on body impedance and the duration of exposure, a relationship charted in IEC 60479-1. That document maps body current against exposure time into zones that run from imperceptible, through muscular contraction and inability to let go, to a high probability of ventricular fibrillation. Body impedance falls as voltage rises and varies with skin moisture and contact area, so the same touch voltage produces different currents in different circumstances. Because the heart is most vulnerable to fibrillation during a specific portion of its cycle, both the magnitude and the duration of current matter. This is why fast disconnection is so important: a current that is survivable for tens of milliseconds may be lethal if sustained for a full second.

Installation rules distill this physiology into a conventional touch-voltage limit that may persist indefinitely without automatic disconnection, widely taken as fifty volts alternating current in ordinary dry conditions and lower in wet or restrictive locations such as bathrooms, swimming pools, and agricultural buildings. Any touch voltage above that limit must be removed within the tabulated disconnection time. Bonding lowers the touch voltage directly, while low loop impedance shortens the time for which it exists; a compliant design uses both.

Protective Earth and Equipment Grounding Conductors

The protective earth conductor, designated PE and identified by green-and-yellow insulation in IEC practice, connects the exposed conductive parts of equipment to the main earthing terminal. In North American terminology, the equivalent conductor is the equipment grounding conductor. Its sole purpose is safety: it carries no current in normal operation and exists to provide a fault path and to hold accessible metalwork near earth potential. Because it must carry potentially large fault currents long enough for protection to operate, its cross-section is coordinated with that of the live conductors: IEC 60364-5-54 permits either a simple table, in which the protective conductor matches the line conductor up to sixteen square millimeters, holds at sixteen square millimeters for line conductors between sixteen and thirty-five, and thereafter equals half the line conductor area, or an adiabatic calculation derived from the fault current, the clearing time, and the conductor material. The green-and-yellow bicolor combination is reserved exclusively for this function, so it must never be used for any other conductor.

The equipment grounding conductor must be continuous and reliable from each piece of equipment back to the source. Connections use dedicated terminals, lugs, or grounding screws rather than relying on incidental metal-to-metal contact. Painted or anodized surfaces are cleaned or fitted with star washers that bite through coatings to ensure a low-resistance joint. The integrity of this conductor is verified during installation by measuring continuity and, where required, the resistance of the connection.

The Grounded Conductor and the Neutral

The grounded conductor, commonly the neutral in single-phase and three-phase systems, is a current-carrying conductor that is intentionally connected to earth at the source. It is distinct from the protective conductor even though both may originate at the same point. The neutral returns load current in normal operation; the protective conductor does not. Keeping these functions separate downstream of the source is essential, because a protective conductor carrying load current would place a voltage on equipment enclosures and defeat its safety purpose.

The point at which the neutral and protective functions are bonded is critical. In a typical service, the main bonding jumper connects the grounded conductor to the equipment grounding system and to the grounding electrode at exactly one location, usually at the main service disconnect. Downstream of this point the neutral and protective conductors are kept separate. Multiple neutral-to-ground bonds create parallel paths that allow normal neutral current to flow on equipment grounding conductors and metal enclosures, a hazardous and noisy condition.

Bonding Conductors and the Main Earthing Terminal

Bonding conductors interconnect extraneous conductive parts, such as metal water and gas pipes, structural steel, and cable trays, with the equipment grounding system. Main protective bonding ties these large metallic objects to the main earthing terminal so that the entire installation forms an equipotential zone. Supplementary bonding adds local connections in areas of increased risk, such as bathrooms, where simultaneous contact with pipework and equipment is likely.

The main earthing terminal, or main grounding busbar, is the central node where the protective conductors, bonding conductors, and the grounding-electrode conductor meet. Concentrating these connections at a single, accessible busbar simplifies inspection and guarantees that every protective path shares a common reference. The grounding-electrode conductor then links this busbar to the earth electrode itself.

Grounding Electrodes

The grounding electrode provides the physical interface to earth. Common electrodes include driven rods, buried plates, ground rings, and connections to the steel reinforcement of concrete foundations, the last known as a concrete-encased electrode or Ufer ground. The objective is a low and stable electrode resistance, which depends on soil resistivity, electrode geometry, depth, and moisture. Several electrodes bonded together usually achieve a lower resistance than any single rod.

Soil resistivity varies enormously, from a few ohm-meters in moist clay to thousands of ohm-meters in dry sand or rock. Because resistance to earth is dominated by the shell of soil immediately surrounding the electrode, increasing length and using multiple, well-spaced electrodes is generally more effective than increasing diameter. Two rods driven close together barely improve on one, since their resistance areas overlap; the usual guidance is to separate rods by at least their driven depth.

Codes set concrete floors rather than a single universal target. The National Electrical Code, for example, requires a single rod, pipe, or plate electrode to be supplemented by an additional electrode unless it is shown to have a resistance to earth of twenty-five ohms or less, and it requires supplemental rods to stand at least six feet apart. A concrete-encased electrode qualifies with at least twenty feet of bare copper conductor no smaller than 4 AWG, or twenty feet of reinforcing bar at least half an inch in diameter, embedded in a foundation footing that is in contact with earth. Because such an electrode borrows the whole footing as its interface with the soil, it usually outperforms a driven rod and does not require supplementing.

Electrode resistance is verified by measurement, classically with the three-point fall-of-potential method, in which a current is injected between the electrode under test and a remote current probe while a potential probe is moved along the line between them; the reading taken at roughly sixty-two percent of the distance approximates the true resistance. Clamp-on testers offer a faster check where the electrode forms part of a multiply grounded loop. Measurements are repeated over the life of the installation, because drought, frost, and corrosion all raise electrode resistance.

TN Systems

In a TN system the source neutral point is directly earthed, and the exposed conductive parts of the installation are connected to that point by protective conductors rather than by a separate local electrode. A fault to an enclosure therefore creates a low-impedance metallic loop back to the source, producing a large fault current that operates overcurrent protection quickly. TN systems are common where the supply and the installation are closely coupled, as in many distribution networks.

TT Systems

In a TT system the source neutral is earthed at the supply, but the installation's exposed conductive parts are connected to a separate, independent earth electrode local to the installation. The fault loop therefore passes through two earth electrodes and the intervening soil, giving a relatively high loop impedance. The fault current is often too small to operate overcurrent devices quickly, so TT installations rely on residual-current devices to detect leakage and disconnect within the required time. TT is common for rural supplies and for installations where a reliable metallic return to the source is unavailable.

IT Systems

In an IT system the source is either isolated from earth or connected through a deliberately high impedance, while the exposed conductive parts are earthed locally. A first fault from a live conductor to earth produces only a very small current limited by the system's capacitance and any inserted impedance, so the installation can continue to operate. This continuity of supply is valuable in hospitals, certain industrial processes, and other settings where an unplanned shutdown is itself hazardous.

The advantage of an IT system comes with an obligation to find and clear the first fault before a second occurs, because a second fault on a different phase would create a high-current path between two energized conductors. Insulation-monitoring devices continuously check the resistance between the live conductors and earth and raise an alarm on the first fault. Skilled personnel then locate and repair it while the system continues to supply critical loads.

Operating Principle

A residual-current device measures the vector sum of the currents in the live conductors passing through it, typically using a current transformer that encircles them. In a healthy circuit the outgoing and returning currents are equal and the sum is zero. If some current escapes to earth, through a person or through faulty insulation, the sum is no longer zero, and the resulting residual current induces a signal in the transformer. When this signal exceeds the device's threshold, a tripping mechanism opens the circuit.

The sensitivity and speed of operation define the level of protection. Devices rated at thirty milliamperes provide what the wiring rules call additional protection against electric shock, because a residual current of this magnitude, interrupted within tens of milliseconds, holds the touch voltage and the shock current below the threshold for ventricular fibrillation in most adults. International practice mandates such devices for socket-outlet circuits rated up to thirty-two amperes that ordinary persons may use, and for mobile equipment used outdoors. Devices with higher ratings, such as one hundred or three hundred milliamperes, serve fire protection and equipment protection, where the concern is energy released into faulty insulation rather than direct contact. A device of any rating is, however, only supplementary to correct earthing and bonding, never a substitute for them.

North American practice sets a tighter threshold for personnel protection. A Class A ground-fault circuit interrupter, defined by UL 943, must trip at a ground-fault current of six milliamperes and must not trip below four milliamperes, and the National Electrical Code requires such devices at receptacles in bathrooms, kitchens, garages, outdoors, and other damp or grounded locations. Ground-fault protection of equipment is a separate function set at much higher levels, intended to limit damage from arcing faults on large services rather than to protect a person. Recognizing which function a device performs is essential, because a device rated for equipment protection offers no meaningful protection against electric shock.

Device Types and Selectivity

Residual-current devices are classified by the waveforms they can detect. IEC 61008-1 and IEC 61009-1, which cover residual-current circuit breakers with and without integral overcurrent protection, define the two simpler classes; IEC 62423 adds requirements for the two more capable ones on top of them. Type AC responds only to sinusoidal alternating residual currents. Type A additionally responds to pulsating direct currents that arise downstream of the rectifiers found in many modern electronic loads, and it is the minimum type now required for ordinary socket-outlet circuits in several jurisdictions. Type F adds tolerance of the composite, multi-frequency residual currents produced by single-phase variable-speed drives. Type B extends detection to smooth direct residual currents and to alternating residual currents up to one kilohertz. It is required where equipment such as three-phase variable-speed drives, transformerless photovoltaic inverters, or electric-vehicle chargers can produce a direct-current component that would magnetically saturate the sensing core and blind a simpler device. Selection follows the load: fitting a Type AC device downstream of a rectifying load is a common and dangerous error, because a modest direct residual current can leave the device unable to trip on a subsequent alternating fault.

In larger installations, selectivity ensures that a fault trips only the device nearest to it, leaving the rest of the installation energized. Time-delayed devices, marked as selective types, are placed upstream of instantaneous devices so that the downstream device clears a fault before the upstream one reacts. Coordinating both the current thresholds and the time delays of devices in series prevents unnecessary loss of supply to healthy circuits.

Nuisance Tripping and Leakage Management

Electronic equipment contains filters that deliberately pass a small current to earth, and many devices on one circuit can accumulate enough standing leakage to approach a residual-current device's threshold. The result is nuisance tripping that has no relation to a genuine fault. Designers manage this by distributing loads across several residual-current devices, by selecting devices with appropriate sensitivity, and by accounting for the cumulative protective-conductor current of connected equipment. A common design rule keeps the expected standing leakage below roughly thirty percent of the device's rated residual current, since a device may legitimately trip anywhere between half and the whole of its rating. Information-technology equipment declares its protective-conductor current in the manufacturer's data, which makes the budget calculable rather than guesswork. Where the leakage is inherent and large, as with banks of switching power supplies or long variable-speed-drive cables, the answer is a separate circuit or a device of higher rating for fire protection rather than a defeated protective device.

Lightning Protection Earthing

A lightning protection system intercepts a strike with air terminals, conducts the current through low-impedance down conductors, and disperses it into the earth through a dedicated grounding electrode arrangement. Because a lightning current rises in microseconds, the relevant property is impedance, not merely resistance: conductor inductance dominates. A straight conductor has an inductance on the order of one microhenry per meter, so a current rising at tens of kiloamperes per microsecond develops tens of kilovolts across every meter of lead. Down conductors are therefore kept short, straight, and generously spaced around the perimeter, and tight bends are avoided because a loop presents far more impedance than its length alone suggests. Multiple down conductors and a ring electrode share the current and reduce the impedance presented to the strike. The IEC 62305 series and NFPA 780 are the principal documents governing this design, and both classify structures by protection level so that air-termination geometry, down-conductor spacing, and earth-termination arrangements scale with the assessed risk.

Bonding is essential to lightning safety. All major metallic systems entering a structure are bonded to the lightning grounding system so that they rise and fall in potential together, preventing destructive side-flashes between, for example, the lightning down conductor and nearby plumbing. Where direct bonding is not acceptable, such as for gas lines, surge-protective bonding through spark gaps maintains equipotential during a strike while preserving normal isolation.

Surge Protective Devices and Reference Grounding

Surge-protective devices divert transient energy to the grounding system, clamping the voltage that reaches sensitive equipment. Their effectiveness depends on a short, low-inductance connection to ground, because the voltage developed across even a short lead during a fast surge can exceed the protection level of the device itself. Installation guidance accordingly limits the total length of the connecting leads, commonly to half a meter, and favors the so-called V-connection that routes the supply conductors through the device rather than tapping it on a stub.

Coordinated arrangements grade the protection by energy. Type 1 devices, tested with the 10/350-microsecond impulse that represents a direct lightning current, sit at the service entrance of a structure with a lightning protection system. Type 2 devices, tested with the 8/20-microsecond impulse, handle the induced and switching surges that reach distribution boards. Type 3 devices provide the final trim at the equipment outlet and must always be preceded by an upstream stage. The grounding and bonding between the stages is what makes the cascade work, because the voltage-limiting action of each device is measured with respect to the bonding network it discharges into.

For data and communication systems, the bonding network itself is engineered. Telecommunications practice distinguishes an isolated bonding network, in which equipment is referenced at a single point, from a mesh bonding network, in which frames, cable trays, and raised-floor stringers are bonded together into a grid that presents a low impedance at high frequencies. Standards such as ANSI/TIA-607 formalize the busbars and backbone conductors that tie a building's telecommunications spaces into one reference. The principle is the same in every case: protection works only when the protective device, the equipment, and the ground reference share a common, low-impedance bonding network. A surge protector connected to a high-impedance or remote ground offers little protection, because the transient voltage simply appears across the inadequate connection instead of being clamped.

Distinct Purposes

Safety ground exists to protect people and to satisfy regulatory requirements; it must provide a reliable fault path and tolerate large fault currents. Signal ground exists to provide a stable, low-noise reference for circuits and to define the return path for signal currents. A connection that is ideal for safety, such as a long protective conductor shared by many loads, may be poor for signals because it carries noise and develops voltage drops. Conversely, a quiet signal reference is not, by itself, an adequate safety earth.

The two grounds cannot simply be left independent, because safety regulations require accessible metalwork to be earthed, and floating signal references invite both shock hazards and interference. The engineering task is to bond them where safety demands while controlling how, and where, the connection is made so that fault and noise currents do not contaminate the signal reference.

Ground Loops and Single-Point Grounding

A ground loop forms when a signal return shares more than one path to the common reference, allowing circulating currents driven by small potential differences or by magnetic coupling. These currents add unwanted voltages to signals and are a frequent cause of hum and data errors. Single-point grounding addresses the problem by joining signal returns at one node, so that no closed loop exists for circulating current. This approach works well at low frequencies, where the dimensions of the wiring are small compared with the wavelengths of interest.

At high frequencies the single-point ideal breaks down, because the inductance of long return conductors raises their impedance and parasitic capacitance provides additional, unintended paths. High-frequency systems therefore favor a low-impedance ground plane and many short connections to it, accepting multiple paths in exchange for minimal impedance. Practical designs often combine the two philosophies, using single-point grounding for sensitive low-frequency sections and a solid plane for high-frequency sections. The full comparison of single-point, multi-point, and hybrid topologies is a signal-integrity question rather than a safety one and is developed in grounding architecture, with the low-noise instrumentation case in grounding strategies for analog systems and the remedies in ground loop prevention. None of those techniques relaxes the safety requirement that every accessible conductive part stay connected to the protective earth.

Isolation and Reference Separation

Where two systems must exchange signals but cannot share a clean ground, galvanic isolation breaks the direct connection. Transformers, optocouplers, and isolated data couplers transfer the signal while interrupting the path that would otherwise carry fault or noise current between the grounds. Isolation allows each system to retain its own safety earth and signal reference, eliminating the ground loop entirely and, in safety terms, providing an additional barrier between hazardous and accessible circuits.

Application Domains

Residential and commercial installations emphasize protective earthing of all socket outlets and accessible metalwork, main and supplementary equipotential bonding, and residual-current protection of circuits accessible to ordinary persons. Data centers and telecommunication facilities add extensive signal-reference bonding networks and tightly coordinated surge protection, because both uptime and equipment integrity depend on low ground impedance. Industrial plants combine power-system earthing with bonding of large machinery, hazardous-area requirements, and protection against static accumulation.

Specialized environments impose additional rules. Medical locations use supplementary equipotential bonding to hold every conductive surface within reach of a patient at a common potential, and the most critical areas use IT systems with insulation monitoring so that a first fault raises an alarm rather than interrupting life-support equipment. Hazardous locations containing flammable gas or combustible dust require bonding of tanks, piping, road tankers, and transfer hoses so that a charged object cannot discharge as an incendive spark; here the goal is a reliable conductive path to dissipate charge rather than a low resistance to carry fault current. Telecommunications and broadcast sites demand robust lightning grounding because exposed antennas and towers are frequent strike targets, and photovoltaic arrays add module-frame bonding and direct-current fault detection to the conventional alternating-current scheme.

Testing and Inspection

Verification confirms that a grounding and bonding system performs as designed. Earth-electrode resistance is measured to confirm an adequately low connection to ground. Continuity testing of protective and bonding conductors confirms that every accessible part is reliably connected to the main earthing terminal. Loop-impedance measurement verifies that the fault loop is low enough for protective devices to operate within the required disconnection time. Residual-current devices are tested for correct tripping current and time using dedicated instruments; the built-in test button proves only that the mechanism releases, not that the device trips at its rated residual current within the required time. Insulation-resistance testing complements these measurements by revealing the deteriorating insulation that eventually becomes a fault.

Inspection does not end at commissioning. Corrosion, mechanical damage, and modifications degrade grounding over time, so periodic re-testing is part of responsible maintenance. Records of measured values establish a baseline against which later results are compared, revealing slow deterioration before it becomes a hazard. Where automatic monitoring is available, as with insulation-monitoring devices on IT systems, continuous supervision supplements periodic manual testing.