When a measurement or a study shows step and touch voltages above the tolerable limits, the grid no longer matches the real conditions: the ground fault current grew, the clearing time changed, the gravel layer was lost, conductors are broken or stolen, the substation was expanded without extending the grid, or the soil differs from the assumed one. It can also be a poorly made measurement. IEEE 80 defines the calculation and IEEE 81 the measurement.
Symptom
A step and touch voltage measurement, an inspection or an updated study shows that, during a ground fault, the voltages in some areas of the substation or the plant exceed the tolerable ones. Sometimes the finding appears at the perimeter fence, in a recent expansion or next to a remote structure. In the worst case, it is discovered because a person received a shock during a fault.
The grounding grid controls those voltages. During a fault, the fraction of the current that enters the soil through the grid, multiplied by its impedance, produces the ground potential rise (GPR). The grid geometry distributes that potential so that the differences a person can touch or step across stay below the tolerable ones, which IEEE 80 calculates from the clearing time and the surface layer. The grounding design and its verification are adapted to the local code and to the utility’s requirements. The same principles apply to industrial grounding.
Possible causes, from the most to the least likely
| Cause | Why it happens | What gives it away |
|---|---|---|
| Ground fault current higher than the design value | The network grew, generation was connected or the transformer was changed | The current short-circuit study gives more current than the design study |
| Surface layer lost or contaminated | Gravel removed, mixed with soil or covered by vegetation | High touch voltages in areas without gravel |
| Broken, stolen or corroded conductors | Copper theft on downleads and fences, corrosion of joints | Poor continuity between equipment and grid |
| Expansion without extending the grid | New equipment or structures outside the covered area | High voltages in the new area or at its edges |
| Poorly coordinated perimeter fence | Fence bonded to the grid without extending it, or isolated without control | High touch voltage on the fence from outside |
| Longer clearing time | Protection settings that lengthened ground fault clearing | Lower tolerable voltages with the same grid |
| Resistivity different from the design | Layered soil or dry season not considered | Resistivity measurement different from the design |
| Transferred potentials | Cables, pipes, rails or neutrals that carry the potential to other places | Voltage at points far from the grid |
| Poorly made measurement | Auxiliary electrodes too close, interference from other grounds | Results that do not repeat with another method |
What to check first
- Measurement method according to IEEE 81: electrode positions, current injection and separation of the signal from noise. See ground resistance test.
- Design data: fault current, clearing time, resistivity and surface layer assumed, compared with the current ones.
- Area where the high voltages appear: edges, fence, expansions, isolated structures.
- Continuity between each downlead and the grid, and visible condition of downleads, connections and gravel.
- Conductors leaving the substation that can transfer potentials.
- Was the measurement made with the right method and without interference?
- No: Repeat the measurement before concluding
- Yes: Are the high voltages concentrated in one area?
- Yes: Is it an expansion, the fence or a point with damaged conductors?
- Yes: Extend or repair the grid and coordinate the fence
- No: Check the surface layer and the transferred potentials in that area
- No: Did the ground fault current or the clearing time increase?
- Yes: Recalculate the grid with the current values and time
- No: Measure the resistivity and recalculate with the real soil
- Yes: Is it an expansion, the fence or a point with damaged conductors?
How it is confirmed
- Measurement of the grid impedance and of the step and touch voltages with current injection according to IEEE 81, in the critical areas.
- Soil resistivity measurement with the Wenner method in several directions, to check the soil model. You can make a first estimate with our grounding resistance tool.
- Continuity tests between downleads and grid with high current.
- Recalculation according to IEEE 80 with the current fault current, the fraction entering the soil, the clearing time and the real surface layer.
- Sampling excavations to check corrosion when the tests suggest it.
The measurement is made with equipment that injects current into the grid and measures step and touch voltages, such as the Metrel MI 3290 earth analyzer.
How to keep it from happening again
- Review the grid whenever the fault current, the clearing times or the substation area change.
- Extend the grid before installing new equipment or structures.
- Keep the surface layer clean and at the design thickness.
- Protect downleads and fences against theft and repair missing parts immediately.
- Measure periodically the impedance and the continuity, with the same method, so results can be compared.
- Control transferred potentials by isolating or separating the conductors that leave the substation.
Where the tolerable voltages come from
IEEE 80 calculates the tolerable voltage from the current a person can withstand without ventricular fibrillation, which depends on the exposure time. The expressions have this form:
| Voltage | Expression (70 kg person) | 50 kg person |
|---|---|---|
| Touch | (1000 + 1.5 · Cs · ρs) · 0.157 / √ts | same term, with 0.116 instead of 0.157 |
| Step | (1000 + 6 · Cs · ρs) · 0.157 / √ts | same term, with 0.116 instead of 0.157 |
Here 1000 Ω is the body resistance, ρs the resistivity of the surface layer, Cs the reduction factor of that layer and ts the shock duration, taken equal to the fault clearing time. With a high-resistivity surface layer, such as gravel, the tolerable voltage rises; without it, the resistivity term disappears and the limit drops. The IEEE 80 values refer to 70 kg unless the design says 50 kg, and the difference matters when comparing a result with another study. Cs is not constant: it depends on the layer thickness and on the contrast between its resistivity and that of the soil, and the benefit of the layer is small if its resistivity is not clearly higher than the soil’s.
Two consequences of those expressions explain most of the findings:
- Clearing time counts as a square root. A protection setting that doubles the clearing time lowers the tolerable voltage by about 30 %, without touching the grid.
- The surface layer is part of the design. Removing the gravel from a switching area is equivalent to noticeably lowering the tolerable touch voltage.
Standards with a different approach
The European standards follow a different path. IEC 60479-1 describes the effects of current on the body and its impedance. EN 50522 sets, for AC installations above 1 kV, the permissible touch voltage as a function of clearing time. Its values and those of IEEE 80 are not interchangeable, because they start from different body models and different assumptions about footwear and surface layer. A study must state which standard it was made with and keep that standard throughout the calculation and the field verification. The chosen standard, and any local code requirement, must be written in the calculation report.
Ground potential rise and transfer
The current dissipated through the grid produces a ground potential rise, GPR, equal to that current multiplied by the grid impedance. A high GPR is not a problem in itself if step and touch voltages comply, but it is for every conductor leaving the substation. A metal fence, a shielded communications cable, a pipe or the neutral of a low-voltage network feeding the control building can carry that potential to a place where the soil is at another value. In the review, each conductor crossing the perimeter must have an explicit answer: it is isolated, it is bonded, or its touch voltage at the other end is verified to be tolerable.
What not to do
- Do not rely only on the grounding resistance: it does not say whether the voltages are safe.
- Do not bond the fence to the grid without extending the grid beyond the fence or without a study.
- Do not measure with auxiliary electrodes too close to a large grid.
- Do not remove the gravel for works without replacing it with the specified thickness and material.
When to call us
If your grid showed high step or touch voltages, or your substation has changed since it was designed, we redesign or verify the grid according to IEEE 80 as part of primary engineering, using the fault current from the updated short-circuit study. This is remote work, available worldwide; field measurements are done under our testing and commissioning services, in Colombia, or elsewhere on request. We work on medium-, high- and extra-high-voltage substations and in industry. Contact us.
Related guides
- Metrel MI 3290 earth analyzer: GP, GL, GF and GX sets
- Grounding grid
- Ground resistance test
- Industrial grounding
- Short-circuit study
- Lightning shielding
- Zero-sequence current
References
- IEEE 80, Guide for Safety in AC Substation Grounding.
- IEEE 81, Guide for Measuring Earth Resistivity, Ground Impedance, and Earth Surface Potentials of a Grounding System.
- IEC 60479-1, Effects of current on human beings and livestock – Part 1: General aspects.
- EN 50522, Earthing of power installations exceeding 1 kV a.c..
Frequently asked questions
What are step and touch voltages?
During a ground fault, the current entering the soil raises the potential of the grid and of the ground. Touch voltage is the difference between a structure a person touches and the ground where they stand; step voltage is the difference between their feet one meter apart. IEEE 80 calculates the tolerable values according to the clearing time and the surface layer.
Does a low grounding resistance guarantee safe voltages?
No. Safety is defined by the step and touch voltages compared with the tolerable ones, which depend on the current actually injected into the soil, the clearing time, the grid geometry and the surface layer. A grid with low resistance can still have dangerous touch voltages at its edges or at the fence.
Why does a grid that used to comply no longer comply?
Because conditions changed: the network grew and the ground fault current is higher, clearing times were changed, the substation was expanded, gravel was removed, or conductors were stolen or corroded. The original design is valid only for the conditions it was calculated with.