IET logo
 
IET
Decrease font size
Increase font size
Topic Title: TNC-S Fault?
Topic Summary: Would a fault with the PEN desensitise a RCD?
Created On: 12 September 2017 04:14 PM
Status: Read Only
Linear : Threading : Single : Branch
Search Topic Search Topic
Topic Tools Topic Tools
View similar topics View similar topics
View topic in raw text format. Print this topic.
 12 September 2017 04:14 PM
User is offline View Users Profile Print this message



iPelec

Posts: 4
Joined: 14 September 2016

Hi all.


Got a little stumped today on an EICR.
1 RCD covering just sockets - House is 10 years old so not unexpected.
ZE 0.01 (Transformer 30m away so not unexpected but still low)
ZS at socket 0.33 again all ok for now - had a disconnected PE in ring found and fixed, came to RCD test and failed, again not unexpected went back to RCD.

Tested direct to bottom of RCD, disconnected straight away.
Ramp test gave no results just disconnected as if cross connection somewhere.
Removed PE and N from Ring in board, same result.
IR >2G
IR rest of board 64M across most of the other circuits.
Tried RCBO and different RCD same results.
Removed all load N and PE from board essentially testing just the RCD with nothing else connected - same result.

Im now thinking that there is a PEN fault causing flow through the PE system. So going back to TT it and try again.

What are your thoughts? Would this stupid low ZE be a sign of this?
I have seen low results before but never that low.
 12 September 2017 07:20 PM
User is offline View Users Profile Print this message



Alcomax

Posts: 215
Joined: 12 November 2009

No quite sure from statements what actually happened regards RCD test. For example :

"Tested direct to bottom of RCD, disconnected straight away ". Do you mean the RCD operated ?

Are you sure there is no error with test equipment? It may be a luxury, but always helps to have a back up loop tester when odd results present.

The service cable to our house is 13 metres to substation [ I was there when it went in ]. We have a Ze of 0.08 ohms, that works out about right for length. So your 0.01 ohm is odd. What about loop test on LN at incomer....is this also 0.01 ohm?

What type of RCD is it ? [ type AC, A etc ], is the house the only one on the transformer or are there other properties quite close?
 12 September 2017 09:50 PM
User is online View Users Profile Print this message



mapj1

Posts: 9258
Joined: 22 July 2004

was IR L and N together versus PE or versus PE and local earth bonding ?
10milliohm Ze suggests PSSC of 23000 amps, which I doubt, assuming a 100A single phase service, but if the least significant digit is the '1' then it might be almost 0.02, which is still quite low, and then there are the instrument tolerances to consider.

0.33 ohms is almost 250m of 2.5mm T and E, so I'm wondering how far away the socket actually was, or if there is another high resistance join somewhere.
EDIT cobblers.
2.5m of 2.5mm2 single is ~16 milliohms. (as is 1m of 1mmsq, 10m length of 10mmsq etc )So 320 milliohms is there and back 10m in 1m, and 25m in 2,5mmsq

I assume the socket is about 10m of cable from the CU ?

Even so, I'd expect you to be able to trip the RCD.
but... if the circuit CPC to that socket returns via the RCD neutral for some reason, then a Zs test will look OK, but the trip may never see an L-N imbalance. Possible if the socket is a spur and there was a cock up in the junction box.

If you are worried about PEN problems the canary in the coal mine, as it were, is to bring in a terra-firma earth from outside, which need only be a (not insulated!) screw driver in the lawn, or a garden fork and a croc-clip. AC voltage from MET to terra firma offset, and see if it moves much when the kettle is switched on.
More than a couple of volts is worth investigating.

-------------------------
regards Mike


Edited: 12 September 2017 at 09:57 PM by mapj1
 13 September 2017 06:54 AM
User is offline View Users Profile Print this message



iPelec

Posts: 4
Joined: 14 September 2016

Ok ir tests were done with no bonding attached.
I did do a l n ze which wa as around 0.05.
Rcd operates but instantly as if there is a cross connection somewhere.

I have a rod in the van so will pop that in later and see what it gives me.
I also did a ze with just the bondings attached to double check they are connected.
These came back as 0.17 each.
So I could test with main earth out which would test the same as a screwdriver. Then rod it if needed
 13 September 2017 01:09 PM
User is online View Users Profile Print this message



mapj1

Posts: 9258
Joined: 22 July 2004

Hmm am I understanding you right ?
Rcd operates but instantly as if there is a cross connection somewhere.

RCD goes off as soon as power is supplied - i.e. before you use your test gear ?
If so I'd suspect a fault from L or N to true earth, not via the CPC. Hence my question about the IR test - if you only test from cable CPC to L+N it will find a fault to the CPC, but may miss a neutral short to a heating pipe or some other non CPC return path.
If not, what did you mean ?

-------------------------
regards Mike
 20 September 2017 09:43 AM
User is offline View Users Profile Print this message



AJJewsbury

Posts: 15716
Joined: 13 August 2003

Im now thinking that there is a PEN fault causing flow through the PE system. So going back to TT it and try again.

If the circuits downstream of the RCD are sound, then current flowing in the PE system won't affect the RCD at all - modern RCDs (unlike the old Vo ELCBs) only monitor L & N - not PE.
- Andy.
Statistics

New here?


See Also:



FuseTalk Standard Edition v3.2 - © 1999-2017 FuseTalk Inc. All rights reserved.

..