IET logo
Decrease font size
Increase font size
Topic Title: Confused need help
Topic Summary: Strange IR readings on lighting circuit
Created On: 12 January 2013 01:36 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 January 2013 01:36 PM
User is offline View Users Profile Print this message


Posts: 599
Joined: 06 November 2010

Nice simple job for a Friday morning, fit a PIR to two existing lights in an 80's bungalow TNCS supply. Tested new cables I put in - no problems, connected everything together:
Power in, Power out, Switch (supplying PIR), PIR, Two lights.
Tested IR on 250V before connecting PIR and going to CU / lights for final tests.
Live - CPC = 58 Ohm (That is not a misprint)

Dismantle everything (Wago Connectors, not the lever sort. agghhh). Test IR on all cables individually. The power output cable has L-CPC 0.34 MOhm. Everything else L-CPC >99.9 MOhm.

Start testing everything in desperation (kneeling on two joists with face mask which is steaming up glasses and head torch) eventually find the Line of the power output cable to the CPC of the power input cable reading 15 Ohm. (Again not a misprint). (PIR not connected yet)
Put everything back together IR Line - CPC = 35 Ohm.

I repeated all the above tests a number of times as I was having difficulty accepting them.

As it is now getting late and I have most of the house lights off I thought it must have been like this before I got here so I will try the big bang test.

Power up the circuit while monitoring earth leakage for the installation and there may have been an increase but it was less than 1mA. Appliances in house were running making exact measurement difficult.

Test power taken by the lighting circuit with nothing turned on and PIR still not connected = 50 mA rather than the 6A the IR would suggest.

Connect PIR and customer is very happy with way it works but I have to go back to find out what the two problems are: Low IR on L-E of Output power cable and very low IR on CPC of supply cable to line of output cable problem is as well as the possible low IR on the power circuits which may be part of the same problem. I had disconnected the TV amp so it is not that.

Customer thinks shaver sockets may be on lighting circuit but I did not have time to test and they should both be on the output power cable side.

I tested at the CU, while the circuit was still dismantled, and all the lighting circuits gave good IR (Not RCD protected). The power side (RCD protected) gave an IR L-E of 0.32M Ohm (appliances still connected)

Any ideas greatly appreciated as all I can think of is some sort of inductive load as the Megger is testing with DC but why it is across L-CPC on two different parts of the cable I have no idea. While I had the breaker to the lights off the rest of the power was on could a fault elsewhere be causing the problem?

I am going to have to strip the circuit down and test at each JB/light fitting, but what am I looking for?

Any suggestions appreciated.
 12 January 2013 01:46 PM
User is offline View Users Profile Print this message


Posts: 329
Joined: 25 May 2006

Is this installation PME?

I think you are seeing another load on the lighting circuit (possibly the primary coil of a shaver socket). This load will be connected between L& N but you are seeing it between L & E perhaps because the installation has an existing N-E fault or maybe because you still have the installation's main switch ON and you can 'see' the N-E link at the service head.

The fact the you see 15Ohms, or so, but you don't pass the expected 6A is because you are testing with a low voltage DC, not with AC. If the load is the coil of a shaver socket then you will see the DC resistance of its winding which will be much less than the impedance that would be seen by 50Hz AC.
 12 January 2013 03:07 PM
User is offline View Users Profile Print this message


Posts: 599
Joined: 06 November 2010

Thanks for the reply, it is PME.

The output power cable from my PIR connection has an IR L-CPC of 0.34 MOhm but N-CPC of >99.9 MOhm. The input power cable has L-CPC >99.9 MOhm and N-CPC of 0.41 Ohm which I am assuming will be the Earth -Neutral link at the head.

As the output power cable Line should not be connected to the supply anywhere else I am assuming a fault on this somewhere. I agree it could be fault somewhere else giving the PME link reading via the fault especially as work has been done on a new shower room a few years ago which is probably when the board was upgraded to 16th edition. I think I am going to have to turn the power off and test all the circuits as a starter as I am currently thinking I may have more than one fault.
 12 January 2013 03:11 PM
User is offline View Users Profile Print this message


Posts: 7772
Joined: 04 July 2007

You're being an "old woman" KFH! Those readings you got can't possibly be right can they or something would be tripping, You're reading through something that's connected or the N-E as Adrian said. (doesn't only apply to PME though).
If it's all working ok don't go pulling it apart again, those readings you got were not "faults" otherwise the "bang" test would have shown them up! Just button it all up,

 12 January 2013 10:06 PM
User is offline View Users Profile Print this message


Posts: 599
Joined: 06 November 2010


I will accept "Pedantic old sod" :-)

I do not like things that give funny readings as I don't know if there is something in the circuit designed to work like that or if it is a fault either in the appliance or the way it has been wired. I tell my customers what I am like so they know what they are getting and generally they are happy with my approach.

But I am still having difficulty thinking of anything which is going to put an inductive load L-E on a domestic circuit.
 13 January 2013 06:33 PM
User is offline View Users Profile Print this message


Posts: 1118
Joined: 13 August 2003

I am not quite sure of the circuit arrangement that you describe and forgive me if I am telling granny how to suck eggs but where you have a N-E connection of low resistance on any circuit and that might be for legitimate reasons (eg front end emc filter) and if the neutral shares a common bus bar then where equipment is connected between line and neutral, there will be a consequent low value between Line and earth. These common bars can be areal pain for lads like you who are trying to do things right! By the way, the 0.34M is an interesting figure. Often seen it on certain rcbos with loss of neutral protection connected (another pain for the tester).
Let us know the outcome and best of luck.


Lyle Dunn
 13 January 2013 09:38 PM
User is offline View Users Profile Print this message


Posts: 599
Joined: 06 November 2010

Thanks for the suggestions. I am going back this week. I don't think they have any emc filters, very ordinary domestic set up, but I will find out.
No RCBOs only one RCD. I am sure I have had 0.34 ish in the past when old RCDs in circuit.
If I can find the answer I will post here.
 15 January 2013 05:20 PM
User is offline View Users Profile Print this message


Posts: 599
Joined: 06 November 2010

After a few false starts due to lighting circuit being wired as a lollipop ring via a two way switch wire and I started down the wrong leg, I found the problem. A neutral - earth fault in a junction box, the last one I checked, with an unsheathed earth touching neutral. I have had N-E faults in the past but this one threw me.

The low IR on the RCD side of the board disappeared although it was still 0.98M, with appliances connected and turned on so I will accept that.

Many thanks for the contributions.

New here?

See Also:

FuseTalk Standard Edition v3.2 - © 1999-2018 FuseTalk Inc. All rights reserved.