Earth fault loop impedance testing: Ze and Zs
Updated 15 October 2026. Written by Acacio Oliveira, founder of FreeMocks, plumbing and heating engineer (F-Gas registered).
The short answer: a loop impedance test measures Ze at the origin with the earthing conductor disconnected, and Zs at the far end of each circuit, or works it out as Zs = Ze + (R1 + R2). Each Zs must be no higher than the maximum for its protective device, corrected for temperature, which in practice is about 80% of the BS 7671 table value.
The earth fault loop is the path fault current takes from the transformer, out along the line conductor to the fault, back along the protective conductor and earthing, and home to the transformer. Its impedance decides how much current flows in a fault, and so how fast the protective device trips.
The earth fault loop
Guidance Note 3 lists the parts of the loop as: the CPC, the MET and earthing conductor, the metallic return path in a TN system (or the earth itself in TT and IT systems), the transformer's earthed neutral point, the transformer winding, and the line conductor from the source to the fault.
Measuring Ze
- Safely isolate the installation: main switch off and locked, or all circuits isolated.
- Disconnect the main earthing conductor from the main earthing terminal.
- With the supply live, test between the supply line conductor and the earthing conductor.
- Isolate again and reconnect the main earthing conductor before anything else.
Compare the reading with the maximum the distributor declares: typically 0.8 Ω for TN-S and 0.35 Ω for TN-C-S. Where TN-C-S or PME applies, Guidance Note 3 strongly recommends checking for diverted neutral current before you disconnect the earth. If you rely on the distributor's quoted Ze instead of measuring it, you still have to test that the earthing is actually there and working.
Measuring or calculating Zs
Measure at the furthest point of each circuit with a loop tester, or calculate it:
Zs = Ze + (R1 + R2)
Guidance Note 3 prefers the calculated method for final and distribution circuits. For a board that isn't at the origin, use the loop impedance at that board, often written Zdb, in place of Ze. Live Zs measurements are taken with earthing and bonding left connected.
Then compare with the max Zs for the device. Example: a B16 lighting circuit with Ze of 0.28 Ω and R1 + R2 of 0.94 Ω gives Zs = 1.22 Ω. The 80% figure for a B16 is 2.18 Ω, so it passes comfortably.
You can check figures like this with our Zs calculator, and look up limits in the max Zs values tables.
Checking against max Zs: the 0.8 rule
BS 7671 Tables 41.2 to 41.4 give maximum Zs values for conductors at their normal operating temperature. Your test is done on a cold circuit, so the reading must be lower. Appendix 3 sets the check as:
Zs(measured) ≤ 0.8 × Uo × Cmin / Ia
Here Uo is the nominal voltage to earth, Ia is the current that makes the device operate in the required time, and Cmin is a voltage factor of 0.95 for a supply that meets the ESQCR. The 0.8 allows for the cable heating up under load.
Guidance Note 3 also shows a more exact method using the actual R1 + R2. In its worked example, Ze is 0.35 Ω and R1 + R2 is 0.2 Ω at 20 °C. Multiplying R1 + R2 by 1.20 to bring it to 70 °C gives 0.35 + 0.24 = 0.59 Ω, comfortably under the 7.28 Ω table limit it was checked against.
Loop test types on a multifunction tester
| Test type | Test current | When to use it |
|---|---|---|
| Two-wire high current | Usually 2 A to 15 A, for no more than about 40 ms | First choice where there is no RCD or RCBO on the circuit |
| Three-wire no-trip | Very small, typically around 15 mA | Circuits with an RCD or RCBO; takes longer and varies more |
| Two-wire no-trip | Very small | Last resort, for example on some lighting circuits; most prone to error |
Guidance Note 3 lists them in that order of preference: always use the higher current test where you can, because it is the most accurate. A high current test can trip some small circuit-breakers, such as 6 A, as well as RCDs. No-trip testers avoid tripping the RCD either by keeping the current under about 15 mA or by DC biasing the RCD core. For safety, the tester cuts the test short if the earth voltage rises above 50 V, or 25 V where livestock are present.
Typical reasons for a high or unreliable Zs
| Cause | What you see | How to find it |
|---|---|---|
| Loose or high-resistance connection | Zs, and R1 + R2, higher than the cable length suggests | Check terminations at accessories and joints, working back from the far end; older corroded terminals add resistance |
| Circuit longer than designed | R1 + R2 matches the cable but is over the design figure | Work out the expected R1 + R2: 55 m of 2.5/1.5 mm² at 19.51 mΩ/m is about 1.07 Ω at 20 °C |
| Smaller CPC than expected | R2 high for the line size | Flat twin and earth made around 1969 to 1981 can have a 1.0 mm² CPC with 2.5 mm² live cores |
| High Ze | Every circuit reads high by the same amount | Measure Ze at the origin and compare with the distributor's figure |
| Instrument error | Readings change each time you test | Check lead contact, electrical noise, harmonics and RCD uplift; repeat, or use Ze + (R1 + R2) |
| Inverters or batteries on site | Odd or unrepeatable readings | Loop readings may be affected by power converting equipment; follow the inverter guidance |
Accuracy near the transformer
Loop testers lose accuracy at very low readings, for example close to a transformer. A tester should resolve to at least 0.01 Ω, but that is not the same as accuracy. Guidance Note 3 says readings under about 0.1 Ω, or under about 1 Ω on a low current range such as 15 mA, can carry significant errors. Those errors carry straight through into any prospective fault current worked out from the reading.
When a Zs reading is too high
- Check the R1 + R2 reading and look for a loose or high-resistance connection, often at a socket or junction.
- Check the circuit length against the design. A longer run than planned can push Zs over the limit.
- Consider a larger cable, a lower-rated or different-type device, or adding RCD protection, depending on the circuit.
- Where supplementary bonding has been used because Zs couldn't be met, BS 7671 requires the effectiveness of that bonding to be verified.
What to record
- Ze: in the supply characteristics on the EIC or EICR.
- Zs: for each circuit on the schedule of test results, the highest value, normally at the far end.
- Zdb: the loop impedance at each distribution board that isn't at the origin.
- Prospective fault current: the greater of the prospective short-circuit current and the prospective earth fault current.
More on the last item in our prospective fault current guide.
Periodic inspection notes
On an EICR you often can't isolate the whole installation, so loop impedance is measured with earthing and bonding conductors connected. That confirms the conductors are connected, but not their full continuity. On motor circuits, test the loop on the supply side of the isolated motor controlgear, then do a continuity test between the CPC and the motor.
Exam traps
- Ze is measured with the earthing conductor disconnected; Zs is measured with everything connected.
- Compare a cold reading with the corrected limit (about 80%), not the raw table value.
- Zs = Ze + (R1 + R2) uses the R1 + R2 from the far end of the circuit.
- Cmin is 0.95 for a UK public supply.
- No-trip tests are less accurate, so a high current test is preferred where there is no RCD.
Practise for it
Inspection and testing questions make up most of the 2391 papers and turn up in the AM2 knowledge test. Practise them under time with a worked answer for every question.
Frequently asked questions
What is the difference between Ze and Zs?
Ze is the part of the earth fault loop outside the installation, measured at the origin. Zs is the total loop impedance for a circuit, including the circuit's own line and protective conductors. Zs = Ze + (R1 + R2).
How do you measure Zs?
With the circuit live, use a loop impedance tester at the furthest point of the circuit, between line and earth. Or calculate it by adding the measured R1 + R2 to Ze.
How do you test Zs on a circuit with an RCD?
Use the tester's no-trip or low-current loop test, which limits the test current so the RCD doesn't operate. The readings can be a little less stable, so repeat them if in doubt.
Why disconnect the main earthing conductor to measure Ze?
So that parallel paths through main bonding to water and gas pipes don't lower the reading. Ze must be the impedance of the supply's own earth path. Reconnect it immediately afterwards.
What do you compare Zs with?
The maximum Zs for the protective device in BS 7671 Tables 41.2 to 41.4, corrected for temperature, which in practice means about 80% of the table value for a reading taken at normal room temperature.
What is the 0.8 factor in loop impedance testing?
BS 7671 Appendix 3 multiplies the limit by 0.8 to allow for the conductors getting hotter, and so more resistive, when the circuit is carrying load. A reading taken on a cold circuit has to be at or below 0.8 × Uo × Cmin / Ia, where Cmin is 0.95 for a normal UK supply.
Which is better, measuring Zs or calculating it?
Guidance Note 3 prefers calculating Zs from Ze plus the R1 + R2 measured during continuity testing, for both final and distribution circuits. A live measurement is still useful as a check, and is the usual method on a periodic inspection.
Why does my no-trip loop reading jump about?
A no-trip test uses a tiny current, typically around 15 mA, so the voltage drop it measures is small and easily upset by noise and other loads. Guidance Note 3 says readings below about 1 Ω on a 15 mA range can carry significant errors. Repeat the test, or use R1 + R2 plus Ze instead.
Do solar inverters and batteries affect loop readings?
They can. BS 7671 notes that loop impedance readings may be thrown off by power converting equipment such as inverters, and Guidance Note 3 says the same for grid-connected and island-mode inverters.
Related
Sources: BS 7671:2018+A4:2026, Regulations 643.7.1 and 643.7.3 (pages 250 to 252) and Appendix 3 (page 438); IET Guidance Note 3 (2022), pages 57, 59, 67, 85 to 90, 130, 140, 141, 160 and 176 to 178. FreeMocks is independent and not affiliated with the IET or BSI. Always follow safe isolation and use test equipment that meets HSE GS38.
