Loop tests and RCD's

[Richard]

Had an old Fluke for years, which has incidentally been faultless . Company purchased a Megger last year however when doing Internal earth fault loop impedance readings are far higher than expected.

I.e..example

Ze = 0.20ohms
r1 + r2 = 0.15ohms
Zs should be about 0.35ohms however coming back around an 1ohm so obviously not right.

Anyway, after a bit of fiddling established it was the Rcd causing the drastic increase in resistance. However my old fluke does't suffer from this problem. So Megger was sent back for repair and no problem found??

Anyway, a Seaward tester turned up in the stores this week and tried that. Same bloody problem!! Likewise that will be going back.

Am I doing something wrong?? Stick with the old fluke I think!!

Any thoughts? TIA

Edit: I'm sorry I don't know the model numbers off hand

 

[Midwest]

Think your find its to do with 3 wire low current testing. Someone more clever than me will explain it to you. Apparently Megger MFT's are prone to it, although I've not experienced it with mine. You could try 2 wire test, isolating the RCD to verify. Think davesparks uses that test all the time, hopefully he'll be along soon to explain.

 

[alanl]

Basic explanation - the Flukes send very short bursts of high current down the lines, then calculate the result, the other Makes send low current, longer duration bursts down the lines to then claculate the result.

This gives more variation in the result between different makes, as using a low current cannot accurately measure the earth loop.
Try it on a non-rcd protected source on both high and low current tests, and the results will very likely not be the same from the same tester.

 

[Richard]

Basic explanation - the Flukes send very short bursts of high current down the lines, then calculate the result, the other Makes send low current, longer duration bursts down the lines to then claculate the result.

This gives more variation in the result between different makes, as using a low current cannot accurately measure the earth loop.
Try it on a non-rcd protected source on both high and low current tests, and the results will very likely not be the same from the same tester.

Thank you for your reply.

Am I right in thinking the high current test trips RCD's? If this is the case, then to get any kind of accurate reading i'm going to have to bypass the RCD each and every time?! :freak:

Not going to happen! Most times, assuming the circuit is good, my colleague and I can basically guess what the Zs will be using the Fluke by judging the run. This gives us confidence when you get a result that it's correct or not.

Sticking with my old tester I think!!

 

[telectrix]

there is a way to avoid bypassing the RCD with links. turn MCB off and lock off. link L-E at the furthest point on the circuit, then measure Zs at the CU. sintra has a step by step guide for this.

 

[Richard Burns]

I drew this diagram to show the above, there are many points within the CU that you can test, just make sure that the circuit you are testing is isolated and remember to remove the remote line earth link!! Top one is normal, bottom one is avoiding tripping the RCD.