Measuring a grounding system has complications that other electrical measurements do not. IEEE 81 points to four main error sources: high-resistance auxiliary electrodes, DC and AC stray currents, inductive coupling between test leads and buried metallic objects. Each has a known way of being reduced.
Short answer. Measuring a grounding system is not like measuring a resistor on a test bench. The soil conducts current by electrolysis, has currents of its own, changes with the weather and is full of metallic objects you cannot see. IEEE Std 81-2012 devotes its clause 6 to these complications. The most frequent errors come from the auxiliary electrodes, stray currents, coupling between the test leads and buried structures.
Why the measurement gets complicated
A real ground system is rarely an isolated electrode in uniform soil. According to clause 6.1, overhead shield wires, buried water pipes, cable shields, adjacent railway tracks and conveyor belts can affect the circuit being measured and introduce significant errors. With urban and industrial growth around substations, it is also increasingly hard to find a suitable place for the probes.
The guide adds two warnings. Shield wires may be isolated on purpose or by poor connections, so a low-voltage test can give a result different from that of a real fault. And to compare with calculated values, the impedance can be measured before interconnecting the external shield wires, metallic pipes and other interferences. Over time, the impedance of a grid usually drops a little as the soil compacts, a year or more after installation.
The error sources
| Source | What happens | How the guide says to reduce it | Clause |
|---|---|---|---|
| High-resistance test electrodes | The test current is insufficient: lower than the instrument sensitivity or comparable to stray currents | Lower the resistance of the current electrode: more depth, water around it, additional rods in parallel | 6.2 |
| High-resistance potential probe | If its resistance is of the order of the instrument input impedance, the reading falls below the real value; it weighs more in sandy or rocky soils | Reduce the resistance of the probe | 6.2 |
| DC stray currents | Cathodic protection of pipelines, DC railways, DC transmission lines and galvanic action between soils and metals alter the reading | Use AC or periodically reversed DC, or pulsed; set a test current well above the stray current | 6.3 and 6.2 |
| AC stray currents | Flow through the soil, the system under test and the test electrodes | Use a frequency not present in the stray current, or a random noise signal; filters or narrow-band instruments | 6.4 |
| Reactive component of a large grid | A grid of less than 1 Ω can have significant reactance | Measure at a frequency slightly above or below power frequency | 6.5 |
| Coupling between test leads | A voltage induced in the potential lead by the current in the current lead adds to the true one | Separate the leads or cross them at 90° | 6.6 |
| Buried metallic objects | Nearby rails and pipes distort the potential contours and increase gradients | Align the probes perpendicular to the direction of the structure and place them as far away as possible | 6.7 |
Stray currents
Soil conducts by electrolysis, so direct current produces chemical reactions and potential differences. Probes, soils of different nature and metals generate galvanic voltages that falsify a DC measurement. That is why periodically reversed or pulsed direct current is used. The guide warns that with that type of signal the values come out close to those of AC, but may not be exact for AC applications. It also mentions that currents induced by solar activity, which can be almost DC, could influence a result.
In AC, most instruments work at frequencies between 50 Hz and 3,400 Hz. Avoiding the power frequency and its harmonics is the basic way for the test signal to stand out from the noise.
Coupling between leads
Inductive coupling between two parallel test leads can reach 0.1 Ω per 100 m at 50 Hz or 60 Hz. It seems small, but a large-area grid has a low impedance and requires long leads to remote earth. In that case the error can be large. The guide sums it up as: negligible in grounds of 10 Ω or more, almost always important at 1 Ω or less, and significant between 1 Ω and 10 Ω.
What it means for the person receiving a report
A ground resistance value without context cannot be judged. It helps to know which method was used, at what distance the probes were placed, at what frequency and current the test was run, and what was buried around it. In large or interconnected grids, a reliable report shows the curve of readings and not a single number. If the result differs a lot from the calculated one, the guide on why calculated and measured resistance differ explains the causes.
For an overview of the test, read the guide on the ground resistance test. For the grounds of an industrial plant, the guide on industrial grounding. The grounding resistance tool is useful for a first calculation with field data, not to replace the measurement. Each method reacts differently to these errors: fall-of-potential tolerates high-resistance probes, clamp-on measurement depends on the signal frequency, and the computer-aided grounding multimeter is designed for sites with a lot of electrical noise. Before measuring in an energized installation, also review the safety risks of the test.
We measure your grounding system
At Tiza Group we measure grounding systems in substations, industrial plants and solar plants, in Colombia, or elsewhere on request, and deliver the interpreted result with the method, distances and test conditions. See the testing and commissioning service or write to us through contact. In the quotation we recommend the test and the equipment your case needs.
Related guides
- Calculated and Measured Earth Resistance: Why They Differ
- Fall-of-Potential: the 62 % Rule Explained
- Clamp-On Ground Measurement, Without Stakes: Limits
- Computer-Aided Grounding Multimeter
- Safety When Testing an Energized Grounding System
References
- IEEE Std 81-2012, IEEE Guide for Measuring Earth Resistivity, Ground Impedance, and Earth Surface Potentials of a Grounding System: clauses 6.1 to 6.7 (general measurement problems) and 8.1.2 (differences between theoretical and measured value).
- IEEE Std 80, IEEE Guide for Safety in AC Substation Grounding.
Frequently asked questions
Why does an earth measurement give different values each time it is repeated?
Because the ground changes with moisture, temperature and soil compaction, and because stray earth currents and buried metallic objects have an influence. That is why the guide recommends, when needed, repeating the measurement and plotting the trend instead of relying on a single reading.
Does putting salt around the probe improve the measurement?
Very little. IEEE 81 states that salt added to the water poured around the test electrodes has minimal value: what matters is moisture. To lower the resistance of an auxiliary electrode, depth, water and parallel rods help more.
From what resistance value does coupling between test leads matter?
The guide gives a rule of thumb: coupling is usually negligible in grounds of 10 Ω or more, almost always important at 1 Ω or less, and can be significant between 1 Ω and 10 Ω. It appears mainly in large, low-impedance grids, which require long leads.