Testing ELI

[Eddy Current]

I have TT earthing and want to check the earth loop impedance because I think the earth rods installed may need replacing, or at least siting elsewhere. Problem is I don't have a tester and, obviously, I need to get one in order to do the job. Because this is just a one-off check of my own property's installation I don't want to spend a fortune on getting an expensive piece of kit - although I would like to be able to check the ELI on a regular basis so I don't want to hire. Can any of you recommend the best way to do this?

I have seen reasonably priced testers for sale that would probably suffice, but they are generally without up-to-date calibration so I guess that would be a problem. I have looked at the Martindale EZ150 tester but it doesn't appear to me to be an adequate test. Anyone here ever used one of these things and compared the results with a more conventional tester?

[John Rumm]

Well you have a few options. The easy way is with an Earth loop tester. You will probably be able to pick up a 16th edition style one from ebay for £50. It won't have the nice tricks like being able to test without tripping an RCD - but more than adequate for what you want, and also has the advantage of making the test at high earth currents, giving a better indication of real world behaviour under fault conditions.

e.g.

The other option is a test using discrete meters and a power supply. The full procedure is described in this article:

I have seen reasonably priced testers for sale that would probably

Calibration on TT is less of an issue. An ohm of error on a TN setup would be a big issue. With an ELI of 10 ohms or more it hardly matters.

You are right, its not. That is really just intended for a quick "is it still basically ok" test and not properly proving a new installation.

[Eddy Current]

Thanks for your very helpful reply.

[Eddy Current]

Would this be suitable?

[John Rumm]

You would need to disconnect all the main equipotential bonding prior to the test so as not to get erroneous readings from parallel earth paths that bypass your main earth rod. Ideally also use a socket near to the consumer unit (although the lions share of the ELI will be Ze in any case with TT)

[Eddy Current]

Thanks again.

I am mainly concerned with doing a Ze test, for which I will switch off the supply at the main incomer and disconnect the earth, just measuring the external loop.

[meow2222]

Obviously you don't, people have been measuring resistance for over a century without them

Really if you have doubts its cheaper to simply sink a 2nd earth rod

NT

[Eddy Current]

without them

Resistance and impedance are not the same. But I take your point it can be done without an earth loop tester, but I'd prefer to use one.

That makes perfect sense, of course, but I'd still prefer to get an accurate reading.

[John Rumm]

The testers typically need power to work... so you can't usually do what you are proposing.

Not sure what leads you get with the Robin one, but on the LT5 it comes with a normal IEC Hot condition kettle style lead. You plug it in, and push the button. It creates a (brief) hefty line to earth short and measures the loop resistance. Hence you get a reading that includes the supply impedance, the cabling from the origin to the socket being tested, and the external earth impedance Ze.

With TT systems the main component of that number will be Ze, and you can generally ignore the rest. The equipotential bonds however have scope to look like additional earth electrodes, and their "fortuitous earthing" could give you a reading significantly lower than just that from the electrode on its own. Since its important that the earthing system functions in isolation (i.e. not relying on any fortuitous earthing from other bonded metalwork), you need to test without those connected. (that also means that you should warn people in the house to stay well clear of anything metal while you do the test!)

[Eddy Current]

Surely the power comes from connecting the tester to the incoming live side of the main switch, which would be turned off to isolate the supply to the installation, and by also connecting the tester to the incoming earth (in this case from the main earth rod) which has also been disconnected. By doing this the installation is taken out of circuit and you are just measuring the impedance of the loop external to the building. Or have I got it completely wrong?

[John Rumm]

No that would be ok as long as you can find a suitable way of hooking it up. My comment was really highlighting that the testers are often setup to make measuring ELI from a socket particularly easy, but that is not quite ideal for measuring Ze of an electrode on its own.

If you can safely connect it up with just line and the earth rod with the rest of the installation out of circuit then you will get the right answer.

[Eddy Current]

Ok, cheers mate. Thanks very much for your help.