Every specification for a building carries a line about the earth: 5 Ω, or 1 Ω, or "as low as practicable". The installer drives a rod, the tester reads 40 Ω, and the argument begins. This guide is about what that number is, why the soil and not the rod decides it, what actually reduces it, and which figure the installation needs, which is rarely the one on the drawing. The earth electrode resistance calculator does the arithmetic for a rod or a line of rods.
What the earth electrode is for
An earth electrode connects the installation's earthing system to the ground itself, and it does several jobs at once. In a TT system, where the consumer has no metallic earth back to the supply transformer, it is the return path for an earth fault: the fault current leaves the live conductor, passes through the faulty appliance to its earthed case, down the protective conductor to the electrode, through the soil, and up the transformer's own earth to the neutral. The current that flows is 230 V divided by the sum of the two electrode resistances and not much else, and during the fault the whole earthed metalwork of the installation sits at a voltage of that current times the consumer's electrode resistance: the touch voltage that a person bridging the metalwork and the ground would receive.
The electrode is also where lightning current goes, whether from a lightning protection system on the roof or from a strike on the incoming line; it is the reference for surge protective devices, which can only clamp a surge to earth if there is an earth to clamp it to; and it holds the installation's earthed metalwork at the potential of the ground around it in normal service, so that a person touching a pipe and a wet floor feels nothing. It is worth separating two ideas that share a word. Equipment earthing (protective earthing) connects exposed metal that could become live to the earthing system so that a fault trips a device. System earthing connects a point of the supply, normally the transformer neutral, to earth, and fixes the voltage of the whole system relative to the ground. The consumer's electrode does the first; the utility's electrode at the transformer does the second; a fault current in a TT system passes through both.
Soil resistivity: the number that decides everything
Soil resistivity, ρ, is the resistance between opposite faces of a one-metre cube of the ground, in ohm-metres. Wet organic soil and marsh are around 10–30 Ω·m; clay 20–100; loam 50–150; sand 200–1 000; gravel 500–5 000; rock and dry stony ground 1 000–10 000. The spread is a thousand to one, and the resistance of any electrode is directly proportional to it, so a rod that gives 40 Ω in loam gives 4 Ω in wet clay and 400 Ω in dry gravel. Nothing about the rod matters as much.
Within one site the value moves with moisture above all: the conduction is by ions in the pore water, and as the soil dries below about fifteen percent moisture the resistivity climbs steeply. It moves with temperature, rising sharply when the ground freezes and somewhat when it bakes; with dissolved salts, which is why coastal and irrigated land reads low; and with depth, since the ground a couple of metres down is usually wetter and more uniform than the topsoil. An electrode measured in August after the monsoon may read two or three times higher in May, and a specification that is meant to be met all year has to be met in the dry season.
Measure it before designing anything that must reach a low value. The Wenner four-pin method (IEEE Std 81) drives four pins in a line at equal spacing a, passes a current through the outer pair, measures the voltage across the inner pair, and gives ρ = 2πaR, where R is the voltage over the current. The result is the average resistivity to a depth of roughly a, so repeating the test at spacings of 1, 2, 4 and 8 m shows whether the ground gets better or worse with depth, which decides whether to drive deep or spread wide.
One rod: the formula and what it says
The resistance of a vertical rod of length L and diameter d in uniform soil is, by the expression derived by Dwight and by Sunde and tabulated in IEEE Std 142 and BS 7430:
Picture the current leaving the rod and spreading into the ground through a series of concentric shells of soil. The shell nearest the rod is thin and small, so its resistance is high; each shell further out has a larger surface for the same current, so its resistance is lower; and the total is the sum of all of them out to infinity. The sum converges because the shells grow faster than the resistance falls, and most of it is close in: for a 2.4 m rod about half the resistance lies within 0.3 m of the rod, about 70 % within a metre, and about 85 % within a distance equal to the rod's own length. The earth "far away" contributes almost nothing; the earth in contact with the rod contributes almost everything.
That geometry explains two things installers find surprising. The rod's diameter barely matters, because it only appears inside the logarithm: going from a 16 mm rod to a 25 mm rod changes ln(8L ÷ d) − 1 from 6.09 to 5.64 and the resistance by 7 %. Buy the rod for its strength and corrosion life, not its cross-section. The rod's length matters a great deal, because it appears outside the logarithm and because a longer rod reaches deeper, and usually wetter, soil that the uniform-soil formula does not even credit: doubling a 2.4 m rod to 4.8 m cuts its resistance by 44 % in uniform soil, and by more where the water table is within reach.
More rods, longer rods, better soil
Because the resistance sits in the shells around the rod, a second rod driven close to the first shares those shells and does not behave as an independent resistor in parallel. Each rod raises the potential of the ground around its neighbour by roughly ρ ÷ (2πs) for a spacing s, and the group resistance of n rods in a line comes out near (R1 + (n − 1) × ρ ÷ (2πs)) ÷ n. At a spacing equal to the rod length the second rod delivers about 85 % of its stand-alone value, at twice the rod length over 90 %; as rods are added the group tends not to zero but to ρ ÷ (2πs), the mutual term itself. The practical rule is a spacing of at least the rod length and preferably twice it, and the practical limit is that rods in a line stop paying after four or six.
Longer rods are usually the better investment: coupled 1.2 m sections can be driven to 6, 9 or 12 m with a hammer, the resistance keeps falling as long as the soil keeps getting wetter, and a single deep rod needs one clamp and one inspection pit. Where the ground is rock or a hard layer stops the hammer, the alternatives are horizontal: a bare copper strip or cable in a trench 0.5–1 m deep, a plate in an excavated pit, or a ring around the building, each with its own formula in BS 7430 and IEEE 142. In an augered hole, a bentonite backfill holds moisture against the rod and lowers the first, most expensive shells for years; conductive cement and proprietary backfills do the same with carbon. Salt and charcoal, still poured into pits on some sites, give a good reading for a season and a corroded rod for ever after.
For a substation or a large plant the answer is a mesh: a grid of buried conductor under the whole site with rods at the nodes, designed to IEEE Std 80 to control step and touch voltages under the full fault current rather than to hit a resistance figure. And for any building with a reinforced-concrete foundation, the reinforcement, bonded at construction, is often a better electrode than anything driven afterwards; IEC 60364-5-54 recognises foundation earth electrodes for that reason.
What resistance you actually need
For shock protection in a TT installation the rule is in IEC 60364-4-41, clause 411.5: RA × IΔn ≤ 50 V, where RA is the resistance of the consumer's electrode and protective conductor and IΔn the rated residual current of the RCD. With a 30 mA RCD that allows 1 667 Ω; with 100 mA, 500 Ω; with 300 mA, 167 Ω. Any rod in any soil meets the first, so the RCD, not the electrode, is what makes a TT installation safe, and the RCD guide is the other half of this one. What the rule quietly requires is that RA be stable: BS 7671, the UK wiring regulations, adds that a value above 200 Ω may not be, because a rod that reads 150 Ω in the wet season can read 600 Ω in the dry one and lose contact altogether as the soil shrinks.
The low figures in specifications come from the electrode's other jobs. A lightning protection system wants a low-impedance path for a steep impulse, and IEC 62305-3 recommends an earth-termination resistance below 10 Ω where practicable. A substation earth must carry the utility's fault current without lethal step and touch voltages, which in practice means 1 Ω or less and a mesh. Telecoms and data centres ask for 1 Ω or better as a clean reference for equipment. A commercial building without any of those is usually specified at 10 Ω or 5 Ω as a proxy for "stable, and low enough that the surge protection works", and that is a reasonable engineering figure; it is not, however, what the RCD needs. Ask what the electrode is for, and design for that.
TN, TT and what Pakistani buildings usually have
The letters describe how the supply and the installation are earthed. In a TN system the transformer neutral is earthed and the consumer is given a metallic protective conductor back to it, either separate (TN-S) or combined with the neutral in the service cable and split at the origin (TN-C-S). An earth fault then sees a low-impedance metallic loop, the fault current is hundreds or thousands of amps, and an MCB clears it in milliseconds. In a TT system the transformer neutral is earthed, but the consumer's only connection to it is through the ground, via their own electrode.
In Pakistan the distribution transformer's neutral is earthed by the utility, and the service to most premises brings phase and neutral only; whatever earth the installation has is a rod the consumer drove. That is a TT system, whatever the drawings call it, and it has a consequence that is easy to miss. With a 40 Ω consumer electrode and a few ohms at the transformer, a dead short from live to the earthed case of a water heater draws about 230 ÷ 45 ≈ 5 A. A 20 A MCB does not notice; the case sits at about 200 V until something burns, and the person who touches it takes the current the MCB would not. An RCD, which trips on 30 mA of imbalance, is the only device in the board that disconnects that fault, which is why TT installations must have one on every circuit that a person can touch, and why RCDs matter more here than in a TN country.
Two habits make TT installations safer. Bond the extraneous metalwork, the water pipes, gas pipe, structural steel and ducting, to the main earthing terminal, so that during a fault everything a person can touch rises together and no voltage appears between a tap and a machine. And never connect the neutral to the earth bar at the consumer's end to "make an earth": that turns the installation into an improvised TN-C-S without the multiple earthing that makes real TN-C-S safe, and a broken service neutral then puts full load current, and full line voltage, onto every earthed case in the building.
The example in numbers
A single copper-bonded rod, 2.4 m by 16 mm, in loam of 100 Ω·m for a small commercial building in Lahore whose specification asks for 5 Ω; 30 mA RCDs on the final circuits.
- ln(8 × 2.4 ÷ 0.016) = ln(1 200) = 7.090; the bracket is 6.090.
- R1 = 100 ÷ (2π × 2.4) × 6.090 = 6.631 × 6.090 = 40.4 Ω.
- Two rods 4.8 m apart: the mutual term is 100 ÷ (2π × 4.8) = 3.32 Ω, so R2 = (40.4 + 3.32) ÷ 2 = 21.9 Ω. Four rods: 12.6 Ω. Twenty-two rods in a 100 m line would be needed for 5 Ω at this spacing, which is not a design.
- One 4.8 m rod instead: 3.316 × (ln(2 400) − 1) = 3.316 × 6.783 = 22.5 Ω, the same as two rods from one hole, and better if the deeper soil is wetter.
- TT check: 50 V ÷ 0.030 A = 1 667 Ω; 40.4 Ω is met many times over. The single rod already makes the installation safe with its RCDs; the 5 Ω is about stability and surges.
The route to 5 Ω here is a Wenner survey to find the wettest corner of the site and a deep-driven rod in a bentonite-filled augered hole there, or the building's foundation reinforcement if it can still be bonded. The calculator reproduces these figures and shows what each change to the rod, the spacing or the soil would do.
Measuring and maintaining it
Measure the electrode after installation with a three-terminal earth tester by the fall-of-potential method of IEEE Std 81: a current spike driven at a distance of several times the electrode's longest dimension (30 to 50 m for a rod), a potential spike moved between the electrode and the current spike, and the reading taken at 62 % of the distance; if the reading changes much as the potential spike is moved a few metres either way, the current spike is too close. The electrode has to be disconnected from the installation for the test, so isolate the supply first: with the installation live, the electrode may be carrying the return of a fault that nobody has noticed yet, and the moment it is unclamped every earthed case is live. A clamp-on tester needs a parallel earth path to work against and is for multiply earthed systems, not for a lone rod.
Record the value and the date, and repeat it in the dry season and then yearly. Look at the clamp when you do: copper-bonded steel, galvanised steel and copper all corrode, faster in acid or salty soils, and the joint corrodes before the rod. Keep the rod head in an inspection pit where it can be seen and tested, use exothermic welds or clamps made for the purpose, and keep the earthing conductor to it continuous and sized for the fault it will carry; the cable fault withstand calculator checks that a conductor survives the fault current for the disconnection time. Ahmedonics designs and commissions earthing for control rooms, plant and substations as part of its custom engineering work, with a resistivity survey first and a measured result last.
References
- IEEE Std 142-2007, IEEE Recommended Practice for Grounding of Industrial and Commercial Power Systems (Green Book)
- IEEE Std 81-2012, IEEE Guide for Measuring Earth Resistivity, Ground Impedance, and Earth Surface Potentials of a Grounding System
- IEEE Std 80-2013, IEEE Guide for Safety in AC Substation Grounding — mesh design for step and touch voltages
- BS 7430:2011+A1:2015, Code of practice for protective earthing of electrical installations
- IEC 60364-4-41:2005, Low-voltage electrical installations — Part 4-41: Protection for safety — Protection against electric shock — clause 411.5, TT systems
- IEC 60364-5-54:2011, Low-voltage electrical installations — Part 5-54: Earthing arrangements and protective conductors
- IEC 62305-3:2010, Protection against lightning — Part 3: Physical damage to structures and life hazard — earth-termination system
- E. D. Sunde, Earth Conduction Effects in Transmission Systems, Van Nostrand, 1949
- H. B. Dwight, "Calculation of resistances to ground", Electrical Engineering, vol. 55, 1936, pp. 1319–1328