The clamp-on ground tester measures the resistance of a loop formed by the electrode, a return path through other grounds in parallel, and the soil. It gives a useful value only if the return has a resistance much lower than the electrode. It does not suit grids with several connections or isolated grounds.
Short answer. The clamp-on ground tester, also called a stakeless tester, measures a resistance without driving probes. One coil of the clamp induces a voltage in the loop and another measures the current that flows. The ratio is the total resistance of the loop. IEEE Std 81-2012 (clauses 8.2.2.5 and 8.2.2.6 and Annex E.2) considers it a quick and useful tool under precise conditions, and warns that it cannot be the only instrument.
How the principle works
The clamp is placed around the down conductor of an electrode. It induces in the grounding system a voltage of a defined frequency, normally between 1 kHz and 3.4 kHz, which drives a current through the electrode, the soil and the other connected grounds. The meter measures that current and shows the voltage-to-current ratio on the display.
The method rests on an assumption: the impedance of the multigrounded return system, not counting the electrode under test, is so small compared with that of the electrode that it can be taken as zero. If that holds, the reading approximates the resistance of the electrode. It is a direct application of Ohm’s law to a circuit that already exists: the meter cannot build the test circuit, because it has no probes.
When it gives a useful value
The typical case is a distribution or transmission line: a pole electrode connected by a neutral or a shield wire to many other grounds in parallel. The neutral closes the loop, and the more returns there are, the smaller the contribution of the other elements to the reading. Even a bad, high-resistance ground among many low-resistance ones is not enough to ruin the measurement. If the returns are few or all of them are high-resistance, the error can be large.
Advantages the guide cites:
- It is quick and easy: there is no need to drive probes or disconnect the electrode from the system.
- It includes in the reading the bonding and the connection resistance of the system. An open or high-resistance bond shows up in the reading, because the grounding conductor is part of the loop.
- It can measure the leakage current that flows through the system, data that fall of potential does not give.
Fall of potential measures only the electrode, not the bonding of the system. That is why the guide considers the two tools complementary: the clamp quickly detects problems and fall of potential confirms them.
When it does not work
Clause 8.2.2.5 lists the limitations:
- It can only be applied to an electrode connected to a relatively low-impedance grounding system.
- If the inductive reactance of the neutral or shield wire is significant compared with the resistance being measured, the error can be large. It weighs more with clamps that work at 1 kHz to 3.4 kHz.
- Corroded splices or connections in the neutral or shield wire alter the reading, although the meter usually indicates an open wire.
- It is not applicable to systems with several connections, such as a substation grid or the grounds of multiply connected poles and structures.
- High-frequency noise has an influence, and with a high-resistance ground the noise-to-signal ratio can be high.
Annex E.2 adds three cases. It cannot be used on isolated grounds, because there is no return path; for that reason it also does not serve to verify a new installation or for commissioning tests. It cannot be used if a lower-resistance return exists through the soil, as at cell towers or substations. And if another part of the grounding system lies within the resistance area of the electrode under test, the reading is lower than the true one, which can give a false sense of security.
- Is there a return through other grounds in parallel, with a resistance much lower than that of the electrode?
- Yes: Is it an individual electrode, with no other ground connections of its own?
- Yes: The clamp can give an approximate value of the resistance, to be confirmed with fall of potential
- No: Not applicable: it is a system with several connections, such as a substation grid
- No: Not applicable: with no return there is no loop, as with an isolated ground or a new installation
- Yes: Is it an individual electrode, with no other ground connections of its own?
How much the frequency distorts
Annex E.2 includes an example: a shield wire of five 61 m spans (No. 2 ACSR conductor) with a ground at each pole. With a 60 Hz signal, a real 1 Ω ground would read 32 % higher, a 10 Ω one 3 % higher and a 100 Ω one 0.2 % higher. At 2403 Hz, a 1 Ω ground would read 3.5 Ω, that is, 248 % above, and a 10 Ω one would read 14.8 Ω. At 3333 Hz, the 1 Ω one would read 4.4 Ω. The reactance of the return lines grows with frequency, and weighs more the lower the resistance being measured.
The guide adds other limitations. The high frequency may be less representative of a power-frequency fault than the 128 Hz of traditional meters. The method is sensitive to noise from nearby electrical equipment and is less effective with very low grounds. Also, the clamp has no built-in check. In fall of potential the result can be verified by changing the probe spacing; with the clamp the result is accepted on trust in the method.
Combination with fall of potential
When the own resistance of an electrode is needed without disconnecting the shield wires, clause 8.2.2.6 describes a combined method. The probes are placed as in fall of potential, and a split-core current clamp measures the part of the test current that goes to the grounding system. The ratio of the measured voltage to that current gives the own resistance. It is often used on multi-leg transmission towers or on towers with guy wires. On a four-leg tower, the resistance of each leg is measured separately before combining them. High-frequency noise also affects it.
For the overall picture of grounding tests, see the guide on the ground resistance test. An example of fall-of-potential equipment: Metrel MI 3152 and 3155. For noisy interconnected systems, the guide also describes the computer-assisted ground multimeter. The sources of error common to any ground measurement are in Ground measurement errors, and the comparison with the design value in calculated and measured ground resistance.
We measure your grounding system
At Tiza Group we measure grounding systems, in Colombia, or elsewhere on request, with the method each installation allows, and we do not accept a clamp value without checking it against another method when the case calls for it. 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
- Fall-of-potential method and the 62 % rule explained
- Computer-assisted ground multimeter
- Ground measurement errors: stray currents and coupling
- Calculated and measured ground resistance: why they differ
- Metrel MI 3152 and 3155: ground resistance and resistivity
References
- IEEE Std 81-2012, IEEE Guide for Measuring Earth Resistivity, Ground Impedance, and Earth Surface Potentials of a Grounding System: clauses 8.2.2.5 (clamp-on or stakeless measurement) and 8.2.2.6 (fall of potential with a clamp) and Annex E.2.
- IEEE Std 80, IEEE Guide for Safety in AC Substation Grounding.
Frequently asked questions
Does the clamp measure the resistance of the electrode or of the loop?
The whole loop: electrode, return through other grounds, and soil, measured in series. The method assumes that only the electrode under test contributes a significant resistance. If the return is not negligible, the reading includes that part as well.
Can I use the clamp on a substation grid?
No. IEEE 81 states that the method is not applicable to systems with several connections, such as a substation grid or the grounds of interconnected poles and structures. It should also not be used if a lower-resistance return exists through the soil, as happens at cell towers or in substations.
Why can the clamp reading be much higher than the real value?
Because the clamp works at a high frequency, from 1 kHz to 3.4 kHz, and if the loop has an inductive reactance comparable to the resistance being measured, the reading goes up. The guide illustrates this with an example in which a 1 Ω ground reads as 3.5 Ω at 2403 Hz.