7.5 Step and Touch Potential and Ground Grid Design Criteria (IEEE 80)

Step and Touch Potential and Ground Grid Design Criteria (IEEE 80)

Quick Answer: Level II task 2.1i.2 lists "step potential and touch potential tests" among the required knowledge for ground grid testing, alongside fall-of-potential, two-point, clamp-on, slope, intersecting curves, and earth resistivity. The governing insight: a low grid resistance does not by itself make a substation safe. Safety is determined by whether the voltages a person can actually be exposed to stay below tolerable limits — and a low-resistance grid with poor geometry can still be dangerous.

---

1. The four potentials

When fault current flows into the earth through a grounding grid, the earth surface develops a potential gradient — highest at the injection point, falling with distance. That gradient is where the hazard lives.

Touch is generally more dangerous than step. A hand-to-feet path drives current directly across the chest and the heart; a foot-to-foot path largely does not. Tolerable step voltages are correspondingly higher than tolerable touch voltages.

Transferred potential is the most dangerous of all, because the person is standing at remote earth (essentially zero volts) while touching something at nearly the full GPR. A grid with a 5,000 V GPR presents almost the entire 5,000 V across a person who touches a fence bonded to the grid while standing outside it. This is why substation fences, communication circuits, and metallic piping receive specific isolation or bonding treatment, and it is a favourite exam scenario.

2. Why grid resistance alone is the wrong criterion

The "25 ohm rule" familiar from NEC 250.53 applies to a single made electrode at a service and has nothing to do with substation grid safety. Likewise a substation target such as "1 ohm or less" is a design convention, not a safety proof.

Consider two grids with identical 0.5 Ω resistance. One is a dense mesh with closely spaced conductors and a deep crushed-rock surface layer; the other is a sparse perimeter loop on bare soil. Both have the same GPR for a given fault current. The dense grid keeps the earth surface potential close to the conductor potential everywhere, so mesh and step voltages are small. The sparse grid lets the surface potential sag deeply between conductors, producing large mesh voltages. Same resistance, very different safety.

The correct chain of reasoning is therefore:

1. Determine the fault current flowing into the grid (I_G) — not the total fault current, since some returns via overhead ground wires, neutrals, and cable shields.
2. Compute the GPR.
3. Compute the actual mesh and step voltages from grid geometry, conductor spacing, burial depth, and soil model.
4. Compare against the tolerable step and touch voltages.
5. If actual exceeds tolerable, change the design — more conductors, closer spacing, ground rods, a deeper or higher-resistivity surface layer, or faster fault clearing.

3. Tolerable limits and what moves them

IEEE 80 derives tolerable body current from the Dalziel relationship, in which the tolerable current is inversely proportional to the square root of the shock duration. Two consequences dominate:

• Faster fault clearing raises the tolerable voltage. Protection speed is therefore a grounding design parameter, not merely a protection parameter. Upgrading relaying to clear in 5 cycles instead of 30 can bring a marginal grid into compliance without touching the copper. This linkage between protection and grounding is exactly the kind of cross-domain reasoning Level III and IV questions are written to test.
• Body weight matters. IEEE 80 gives separate criteria for a 50 kg and a 70 kg body; the 50 kg criterion is more conservative and is the usual design basis where the public or lighter personnel may be present.

The surface layer. A layer of high-resistivity crushed rock — typically 4 to 6 inches of clean, washed stone — placed over the substation yard adds resistance in series with the body's foot contact, and this substantially raises both tolerable touch and tolerable step voltage. Two field-critical points:

• The rock must be clean and washed. Fines, dirt, and organic debris that accumulate over years, and especially wet contaminated rock, dramatically reduce its resistivity and erode the safety margin the design depended on. A yard that has not been re-surfaced in decades may no longer meet the design it was built to.
• Depth matters, and a thin or eroded layer does not deliver the assumed derating factor.

Assessing surface layer condition is a legitimate and often-overlooked field observation during a grounding survey.

4. Measuring step and touch potential

Calculations are validated by measurement, typically by injecting current into the grid from a remote electrode and measuring the resulting voltages, then scaling the results to the design fault current.

• Touch measurement: a voltmeter between the structure under test and a probe placed in the earth 1 metre from the structure, with a resistor (commonly 1,000 Ω) in the measuring circuit to represent body resistance.
• Step measurement: two probes 1 metre apart on the earth surface, again through a body-representing resistance.
• Scaling: the measured voltage per ampere of injected current is multiplied by the design grid current. Because injection is typically tens of amperes against a design fault of thousands, the scaling factor is large and any measurement error scales with it.
• Interference rejection: substation yards carry substantial stray 60 Hz current, so measurements use an offset frequency or a filtered instrument, exactly as in fall-of-potential testing.

Locations to test: operating handles of switches and breakers, fences (particularly at gates and corners, and where a fence leaves the grid perimeter), transformer and equipment enclosures, control building doors, and the perimeter where the gradient is steepest.

5. Common grid deficiencies found in the field

• Corroded or broken connections. Exothermic (Cadweld) connections are the standard for buried joints because bolted connections corrode; a bolted buried joint is a deficiency.
• Equipment risers disconnected or theft-cut. Copper theft has become a routine finding on unmanned sites.
• Fence bonding incorrect. Either unbonded where the design requires bonding, or bonded where it creates a transferred potential path to the outside.
• Additions not bonded. New equipment installed without a riser to the grid.
• Surface layer degraded, as above.
• Soil resistivity changed — a site model built on a dry-season survey may not represent seasonal extremes, and the design should use the worst applicable case.

Exam trap: A question states that a substation grid measures 0.4 Ω and asks whether the grounding system is adequate. Grid resistance alone answers nothing about safety. The question that determines adequacy is whether the calculated step and touch voltages stay below the tolerable limits for the design fault current, the clearing time, and the surface layer in place.