I am currently doing a level 3 electrical course. we are having to do a portfolio as we go along. the electrician i have been put wit isn't very helpful. i have done most of it but i am struggling with the fault finding. i just wondered whether anyone could help me with examples of electrical faults in domestic properties and the steps to diagnosing and rectifying these faults. thanks in advance.
fault finding
- Thread starter justjess
- Start date
Help Support Talk Electrician Forum:
- Joined
- Nov 28, 2009
- Messages
- 14,779
- Reaction score
- 966
Let's have a try, but rather than just give you the answers, lets make you do some thinking so you understand what you are doing.
So you get called to a house, because the rcd keeps tripping. It's a modern split load dual rcd board and the left hand rcd is tripping as soon as you try to reset it. Some basic tests show that it only trips when the "downstairs lights" mcb is turned on. With that turned off it resets and stays reset and everything else appears to work.
From your experience so far, what do you think might cause that, and how would you start testing that?
P.S I chose this example because this is exactly the fault I had on this morning's job.
So you get called to a house, because the rcd keeps tripping. It's a modern split load dual rcd board and the left hand rcd is tripping as soon as you try to reset it. Some basic tests show that it only trips when the "downstairs lights" mcb is turned on. With that turned off it resets and stays reset and everything else appears to work.
From your experience so far, what do you think might cause that, and how would you start testing that?
P.S I chose this example because this is exactly the fault I had on this morning's job.
Last edited by a moderator:
possibly a short circuit? wouldnt this be picked up on an insulation resistance test? it could be picked up on a continuity test however as its a short circuit there may be an alternative return path ... or have i got that mixed up?
Are the downstairs lights fed from the tripping RCD?
I realize this is a hypothetical question, but that would be spotted straight away and could indicate a different fault to test for.
I realize this is a hypothetical question, but that would be spotted straight away and could indicate a different fault to test for.
- Joined
- Nov 28, 2009
- Messages
- 14,779
- Reaction score
- 966
It's not a short circuit, that would probably trip the mcb, but that was not tripping. An rcd trips because the outgoing current in the L does not match the return current in the N, so some current is going elsewhere.
So insulation testing is what you need, so how would you do that, what would you measure, from where to where?
P.S to Rob, yes the downstairs lights circuit is on the rcd that is tripping (otherwise that would be a sneaky one to throw at a novice)
ahh yeah i was thinking about the mcb ... well i have a fluke multi-fuctional tester so it would go onto the insulation setting at 500volts and would be measured in megaohms.... then measure between the l-n, n-e, l-e..... the perfect reading would be ≥500 so the higher reading the better
- Joined
- Nov 28, 2009
- Messages
- 14,779
- Reaction score
- 966
What, if any, cables would you have disconnected first?
So you do those tests and it shows 0.05 Megohms between L and E and also between N and E What next?
So you do those tests and it shows 0.05 Megohms between L and E and also between N and E What next?
wouldnt you remove the cables for that circuit so that no other circuits can affect the results and use the croc clips to test with ?
so that needs investigating as its below 2 megaohms and it would show that there is an earth fault somewhere on the circuit which means there may be a break or some sort of damage to the earth cable?
so that needs investigating as its below 2 megaohms and it would show that there is an earth fault somewhere on the circuit which means there may be a break or some sort of damage to the earth cable?
- Joined
- Nov 28, 2009
- Messages
- 14,779
- Reaction score
- 966
That's right, you definitely want to remove the cable(s) of that circuit from the CU so you know you are only testing that circuit.
It won't be a break in the earth cable. There is some fault that is causing excessive leakage to earth, and you have to find that. It could be anywhere in the entire downstairs lighting circuit, so you need a logical way to find where it is.
This is where you need to understand how it's wired, e.g loop at light, loop at switch, etc so know where to split circuits. This one was wired loop at switch, which is very common up here. So how might you pin down where the fault is?
- Joined
- Nov 28, 2009
- Messages
- 14,779
- Reaction score
- 966
Sorry I have to go out just now. I will be back later to pick this up again.
here we do the loop into the lights...im guessing you would leave the cables disconnected from the consumer unit end safely and then go to each switch on the circuit doing an insulation test until you get a rubbish reading and it will tell you that the fault is likely to be on the bit of cable going from the last switch you tested to the one you just tested if that makes sense.
- Joined
- Nov 28, 2009
- Messages
- 14,779
- Reaction score
- 966
If you just do an IR test at each point you would still see the fault. you need to split the L's and N's wherever they loop, in your case at the light fitting, and then test each leg of the circuit
bluetobits
Member
- Joined
- May 10, 2010
- Messages
- 294
- Reaction score
- 6
Or half split technique, then half, half again etc...
A perfect reading is not over 500 meg ohms ........ Rarely do you see figures this high on older properties.
Above 2 is acceptable, .......
Risteard
Senior Member
Well, it should be remembered that the cpc must not be disconnected so that it is connected to the earthing arrangement when the test is being carried out. (In fact this was one of the new requirements when the 17th Edition came out in 2008.)
It should be remembered that the 2 Megohms and other minimum insulation resistance readings quoted in BS7671 for new installations are not for an individual circuit but for the entire installation (or, in larger installations, a distribution board with all final circuits) measured in parallel.
Similar threads
