The minimum size of a grounding grid conductor is set by the heating that the fault current produces during the clearing time: cross-section, current and duration are related by a thermal equation and the material constants. In practice a larger size is chosen for mechanical strength, corrosion, voltage drop, maximum allowed temperature and future growth.
Short answer
The conductor of a grounding grid must withstand, without melting or deteriorating, the worst combination of fault current and duration over the whole life of the installation. IEEE Std 80-2013 relates conductor cross-section to current, time and material constants through a thermal equation. The resulting size is a minimum: a larger one is almost always chosen for mechanical reasons, corrosion and future growth. Where the design current comes from is covered in maximum grid current and split factor.
What each grid element must meet
The standard requires every part of the grounding system (grid conductors, connections, ground leads and main electrodes) to meet the following over its expected service life:
- Enough conductivity, so it does not add significant local voltage differences.
- Resistance to fusing and mechanical deterioration under the worst combination of magnitude and fault duration.
- High reliability and mechanical robustness.
- Ability to keep working even if it suffers corrosion or physical damage.
How current, time and cross-section relate
The standard models conductor heating with a thermal equation that evaluates the current capacity of any conductor whose material constants are known. It starts from two assumptions:
- All the heat stays inside the conductor, with none dissipated to the surroundings (adiabatic process).
- The product of specific heat and density, called thermal capacity per unit volume, is taken as nearly constant. The assumption holds over a wide temperature range as long as the fault lasts a few seconds.
For annealed copper, the standard gives a simplified form: the cross-section in kcmil is the current in kA multiplied by a material constant (7.00 for soft copper, with a 40 °C ambient and a limit temperature equal to the melting point) and by the square root of the time in seconds. With it, the standard works out a 20 kA, 3 s fault:
| Material | Result | Size used |
|---|---|---|
| Soft copper | 242.5 kcmil | 250 kcmil |
| Copper-clad steel, 40 % conductivity | 362 kcmil | 19-strand No. 7 cable |
| Steel | 552.5 kcmil | 3/4 in diameter conductor |
The same equation can compare a fixed size at different times. A soft copper 4/0 AWG (211.6 kcmil) withstands about 42.7 kA for 0.5 s and about 17.5 kA for 3 s: the longer the duration, the lower the allowable current.
IEEE Std 837 includes these same equations in its Annex B.
- Maximum fault current the conductor will carry. Includes future growth of the system
- Maximum clearing time. With backup protection
- Material and limit temperature. Material constants and melting or design limit temperature
- Minimum thermal cross-section. Result of the standard's equation
- Selected size. Increased for mechanical strength, corrosion and ground leads
Material constants
Table 1 of the standard gathers the constants of the most common materials. Some data:
| Material | Conductivity (% IACS) | Melting temperature | Thermal capacity (J/cm³·°C) |
|---|---|---|---|
| Annealed soft copper | 100 | 1,083 °C | 3.4 |
| Commercial hard-drawn copper | 97 | 1,084 °C | 3.4 |
| Copper-clad steel, wire | 40 or 30 | 1,084 °C | 3.8 |
| Steel 1020 | 10.8 | 1,510 °C | 3.8 |
| Stainless steel 304 | 2.4 | 1,400 °C | 4.0 |
| Galvanized steel (rod) | 8.6 | 419 °C | 3.9 |
In bimetallic materials, the melting temperature that counts is that of the metal with the lower melting point: the conductor’s integrity must be kept throughout the life of the substation. For copper-clad steel, independent tests show that the real fusing current can differ from the calculated one, because the specific heat of steel varies widely between 550 °C and 800 °C. Manufacturers can supply test data for their conductors.
Asymmetrical currents
The basic thermal equations are for symmetrical current, with no dc component. When that component is to be considered there are two ways:
- Use the decrement factor. An equivalent symmetrical current is calculated, always larger than the real symmetrical one, because it assumes the ac component does not decay.
- Use the asymmetrical current tables. The dc component heats the conductor more and makes the mechanical forces and absorbed energy almost four times those of an equivalent symmetrical case. Tables 3 to 6 of the standard give the limit currents for copper with different degrees of dc component, obtained theoretically and verified by tests.
The effect of the dc component can be neglected if the current lasts 1 s or more, or if the X/R ratio at the fault point is below 5.
Why a larger size is chosen
The size from the equation is a floor. The standard lists the reasons to go up:
- Mechanical strength and corrosion. The conductor must withstand the expected mechanical and corrosive abuse. In soils that promote corrosion a larger size may be advisable, even when conductor and connection have passed the IEEE 837 tests.
- Conductance. It must avoid any dangerous voltage drop during a fault, throughout the life of the installation.
- Maximum temperature. Near flammable materials the temperature limits are stricter. If the strength of hard-drawn copper is needed, it is best not to exceed 250 °C so it is not annealed.
- Safety factor, as with any electrical component.
- Future growth. It is cheaper to include the margin in the initial design than to reinforce ground leads later.
- Equipment ground leads. They may carry all the current entering the grid; for that reason they may need a larger size or multiple conductors. Lightning current rarely requires more: the size chosen for the fault is usually enough for short-duration surges.
In practice, mechanical reliability sets the minimum size.
How we do it at Tiza Group
Current, backup time and material come from your short-circuit study and protection coordination. At Tiza Group we verify the size within the grid design for medium, high and extra-high voltage substations. You can make a first check with the free Grounding resistance tool. Learn about our engineering design and studies service or write to us. We deliver it remotely, worldwide.
Related guides
- IEEE 80: Safety in Substation Grounding
- Decrement Factor Df and Fault Asymmetry
- Grounding Grid Materials and Connections
- Substation Grounding Grid: Design and Testing
- Grounding of Distribution Substations: Design and Tests
References
- IEEE Std 80-2013, IEEE Guide for Safety in AC Substation Grounding, clauses 11.1 and 11.3, equations (37) to (47); Tables 1 and 3 to 6.
- IEEE Std 837, Standard for Qualifying Permanent Connections Used in Substation Grounding.
Frequently asked questions
What clearing time is used to size the grid conductor?
The longest possible clearing time, backup included. The standard explains that a relay failure can stretch clearing beyond the primary: in small substations backup can reach 3 s or more, while in large ones, which have redundant protection, the fault is usually cleared in 1 s or less.
Can equipment ground leads be the same size as the grid conductors?
Not necessarily. A ground lead from an equipment item may carry the whole fault current entering the grid, while each grid segment carries only a fraction. For that reason ground leads may need a larger size or several conductors in parallel.
Why isn't the size from the thermal equation enough?
Because the equation only prevents the conductor from melting. Practice also requires mechanical strength, enough conductance to avoid dangerous voltage drops, temperature limits near flammable materials, a safety factor, and a margin for corrosion and for future growth of the system.