![\"\"](\"https:\/\/d2z8ufzpcqvblm.cloudfront.net\/sites\/www.voltimum.co.uk\/files\/portal\/DanArticles\/muk648_-_meggers_new_mft1741_main.jpg\")
<\/p>\nA common complaint about earth loop impedance testing is that if several tests are carried out on the same circuit, it often happens that each test gives a different result. Why is this and what can be done about it?<\/p>\n
Earth loop impedance testing is,\u00a0without doubt, the topic that generates the largest number of calls to the Megger technical support service. And the most common question is why results vary so much. It seems logical to expect that if two measurements are made one after the other on the same circuit, the results should be identical or at least very similar. But sometimes they\u2019re not, and in these cases, electricians often suspect that there\u2019s something wrong with their tester.<\/p>\n
Of course, that might be true, but it\u2019s actually very unlikely. So what is going on? In reality, the most likely culprits are noise and harmonics in the supply system, so let\u2019s look at these issues in more detail.<\/p>\n
Loop impedance testing typically works by monitoring the voltage between the live conductors or between the live and\u00a0PE (earth) conductors in a circuit, while a resistor of known value is momentarily connected between them. The change in voltage produced by connecting this resistor is measured and, from this, the instrument calculates the impedance of the \u201cloop\u201d circuit.<\/p>\n
If the resistor is chosen so that the current that flows in it is reasonably large \u2013 around 2.5A or more is often used \u2013 this method works very well and gives accurate, consistent results. This is a high-current two-wire test and it\u2019s very useful, but it has one big drawback. When testing live to earth, the current that flows in the resistor will trip any RCD or RCBO protecting the circuit.<\/p>\n
One solution is to link these devices out before performing the test, but this is time-consuming and potentially hazardous, as the circuit has no earth leakage protection while the links are in place. There is always the possibility of the links accidentally being left in place after testing is completed, and of screws being poorly tightened after the link has been removed. For these reasons, linking out the RCD is very much a last resort and should be avoided.<\/p>\n
Because of the inconvenience of high-current two-wire testing on RCD protected circuits, instrument manufacturers have laboured hard to develop alternative solutions. One approach is to perform the test as described, but choose a resistor that passes a current \u2013 say 15 mA \u2013 that won\u2019t trip the RCD.<\/p>\n
In principle, this works, but the voltage change produced by this small current is tiny and hard to measure accurately. Results can be affected by variations in the supply voltage, load switching, noise in the circuit or harmonic distortion.\u00a0 All of these can produce effects larger than the voltage change the instrument is trying to measure.<\/p>\n
To minimise these problems, instruments that offer no-trip loop impedance testing usually make many measurements over a test period of 15 seconds and, at the end of this time, take an average to produce the result displayed. This approach usually provides good results on circuits with lower levels of noise and harmonics, but it\u2019s not infallible and on \u201cdifficult\u201d circuits, the results may be inconsistent, varying considerably from test to test.<\/p>\n
It\u2019s clear that there can be numerous reasons why loop impedance test results are inconsistent. But does it matter? It may not, because if all of the results are lower than the maximum value prescribed in the IET Wiring Regulations, the circuit can be considered satisfactory.<\/p>\n
There is, however, always the problem of deciding which of the results to record on the test schedule and there are also cases where the variation in the results exceeds the maximum permitted loop impedance. What can be done?<\/p>\n