R1+R2 continuity both ways

[shinypen]

Hi. Testing an old lighting circuit today after a c/u change. First few readings were all fine. Then on one rose I got a reading of 0.5 ohms switch closed (fine) and 186 ohms switch open.

Is this indicative of an insulation problem or something else? I did an IR test and got 0.75 ohms but it was a bit rushed and couldn't be sure it wasn't just an error. That is very low surely? I'll do that again tomorrow.

I'd be grateful for suggestions!

 

[Deleted member 26818]

Could be an insulation problem, or it could be a fluorescent fitting?

 

[shinypen]

Thanks. A fitting wouldn't give a continuity reading both ways though would it? I need to be sure of the ir test first I guess. Also it's rcd protected (tested fine) and that's not tripping.

 

[Wilko]

... I did an IR test and got 0.75 ohms but it was a bit rushed ...

Hi - IR unlikely to be able to read that low, so perhaps repeat that one? Maybe something else connected - smoke alarm or some neon?

 

[Pete999]

Hi. Testing an old lighting circuit today after a c/u change. First few readings were all fine. Then on one rose I got a reading of 0.5 ohms switch closed (fine) and 186 ohms switch open.

Is this indicative of an insulation problem or something else? I did an IR test and got 0.75 ohms but it was a bit rushed and couldn't be sure it wasn't just an error. That is very low surely? I'll do that again tomorrow.

I'd be grateful for suggestions!

IR Test 0.75 ohms are you sure you have that right, it's always best practice to do the testing prior to any CU changes

 

[shinypen]

It's just a lighting circuit so no neon and disconnected the smoke alarm. I did rush it so just closed all light switches and tested ir between line and cpc. IR could well be an erroneous reading but the continuity reading I repeated again and again.

 

[davesparks]

With an IR reading so low you have presumably left the circuit isolated?

Why did you rush the testing?

 

[shinypen]

It's just a lighting circuit so no neon and disconnected the smoke alarm. I did rush it so just closed all light switches and tested ir between line and cpc. IR could well be an erroneous reading but the continuity reading I repeated again and again.

Sure you're right. It's my c/u and the old tiny wylex board was packed full and the frame was falling apart from woodworm! Decided to just change it and pick up the peices afterwards.

 

[Octopus]

What was your Ze ? and Zs at the fitting?

 

[Richard Burns]

If Neutral and earth are connected in the CU bars and the main switch is closed then when the light switch is open the meter could measure a resistance loop, via the supply neutral earth connection, through the load and this could give a high reading of the sort of level you have measured.

 

[telectrix]

i assume the 0.75 ohms IR is a typo and you meant Mohms?

 

[HandySparks]

Did you leave the lamps in?

 

[shinypen]

Ze @ board was 0.2. R1R2 was 0.5. Didn't directly test as didn't want to energise it.

The ir must have been an error and I'll look in to it more tomorrow. Was 0.7 ohms nor m ohms! Most likely I'm just testing a lamp or something although as I say would ir line to cpc pick up this?

The continuity issue I'm sure about (that's what made me jump to ir test) so I'll dig further tomorrow. Just wondered if anyone magiclly had a solution from their experience!

 

[telectrix]

if you has an IR reading of 0.7 ohms, that would equate to 343A. surely that would cause a hell of a bang.

 

[shinypen]

Exactly! It's been running fine prior to my c/u change. I'm sure the continuity reading is right and pretty sure my ir reading is wrong which leaves me confused! Looks like I'll be doing some head scratching and floor boars lifting tomorrow (after I do a very well considered ir test).

 

[Loki]

Just a thought - I know some of the the loft light switches we install on new builds have a neon light on the switch.

Michael

 

[7029 dave]

What cable?, pvc , that is a very low reading, lamps out.?Maybe something else on the circuit.TV booster etc.

 

[westward10]

Am I the only one who has no idea what these readings are all about.

 

[Pete999]

Am I the only one who has no idea what these readings are all about.

No Westy you are not the only one.

 

[7029 dave]

Am I the only one who has no idea what these readings are all about.

I am taking it he means Mohms Westie.?

 

[PEG]

I'm not brave enough,to ask what the readings were,before the board swap...

 

[Anthony_wayne]

No damp areas? Almost like there is something else in the circuit? Or when you IR'd did you have cpc on the bar still with neutral connected?

 

[dmxtothemax]

could there be a switch with built in ilumination ?

 

[Pete999]

I'm not brave enough,to ask what the readings were,before the board swap...

Get the impression that there wasn't any testing before CU change hope I'm wrong

 

[Deleted member 9648]

It always mystifies me when an (alleged) low IR is encountered that a neon always seems to get the blame. This would only happen on a L-N test, the OP has not specified what IR test has been carried out but I doubt it's L-N. Or is it OP?

 

[shinypen]

Tested polarity of incoming supply, ze@ board, direct zs and pfc of several points on system and bonding before hand. All were fine. The board was a mess and a real tangle of too short cables.

I freely admit to being new to this so as an aside is it really common practice to do a full set of tests on every part of every circuit before a board change?

In this instance it's my board and any problems that arise need to be fixed by me anyway. Surely safer and easier once a new board in place.

 

[shinypen]

No damp areas? Almost like there is something else in the circuit? Or when you IR'd did you have cpc on the bar still with neutral connected?

Thanks anthony. This is helpful. I did indeed have the cpc still connected to the bar. Hadn't realised this could introduce problems. I'll update this post at the end of the day with the results! Hopefully just that and an incorrect ir test...

 

[Pete999]

Yes it's always best to IR test before CU change why? because you are adding RCDs to the system, and any IR N-E L-E will trip the RCD, nothing more frustrating than changing CU all nice and neat, then do your testing, no bonding IR faults file:///C:/Users/Pete/Downloads/Best-Practice-Guide-1-Issue-3-(1).pdf

 

[shinypen]

Yeah can see the logic. Point taken. IR tests are such a pain though!

 

[Pete999]

Yeah can see the logic. Point taken. IR tests are such a pain though!

Even worse pain is not IR testing and the RCDs trip and you have to start ripping your nice neat CU to bits to find the fault

 

[David M]

You still need to clarify which conductors you IR tested between to get the 0.7Mohms (assuming this is what you meant). If you do thorough tests then you can start to eliminate the possibilities causing the fault (again assuming there is one). Its good practice to be methodical with your testing rather than just rushing and second guessing.

 

[shinypen]

Thanks everyone. Lesson learnt. My error on both.

The cpc continuity is now fine. I'd left cpc in cu met bar. When I removed it the readings were fine.

again, my error with ir. Reading was as stated by others actually 0.7 m ohms. Looks like I've got an issue with the lighting insulation.

Methodical step by step testing the correct way I completely agree, but hadn't realised leaving cpc in met caused errors which lead me down the wrong path. Thanks.

 

[shinypen]

To clarify. Original ir test was cpc to line. Done all ir test combinations now.

 

[Anthony_wayne]

It's less the leaving the CPC in the main earth bar and more to do with the neutral conductor being left connected. The correct prescribed method is to leave the cpc connected, but remove both line conductors.

 

[David M]

Thanks everyone. Lesson learnt. My error on both.

The cpc continuity is now fine. I'd left cpc in cu met bar. When I removed it the readings were fine.

again, my error with ir. Reading was as stated by others actually 0.7 m ohms. Looks like I've got an issue with the lighting insulation.

Methodical step by step testing the correct way I completely agree, but hadn't realised leaving cpc in met caused errors which lead me down the wrong path. Thanks.

I understand that you are relatively new to the industry (you mentioned this earlier in the thread) Inspection & testing isn't just about taking readings and checking them against what is accepted. Its also about understanding what the tests are for, why the readings are what they are and the implications of such, whether acceptable or not.
Fault finding requires understanding and knowledge of testing, also experience which has to be earned.
I have watched many a spark just go through the motions without any thought to why he/she is doing the tests in the first place.
That's why in my opinion methodical thorough testing and a clear understanding of what you are doing and looking for is vital.
Its great that you are questioning your readings, but its important that those readings are accurate, appropriate and correctly interpreted.

 

[Richard Burns]

It's less the leaving the CPC in the main earth bar and more to do with the neutral conductor being left connected. The correct prescribed method is to leave the cpc connected, but remove both live conductors.

It should be clear that this is the action to take for IR testing and that continuity testing should normally be undertaken with no possibility of parallel paths reducing the reading.

 

[Anthony_wayne]

It should be clear that this is the action to take for IR testing and that continuity testing should normally be undertaken with no possibility of parallel paths reducing the reading.

Agreed mate, though we were talking about his IR test so thought that was clear enough. I see a lot of experienced sparkies still disconnecting the CPC for the IR tests, they don't believe it changed until you show them GN3.

 

[shinypen]

-

 

[shinypen]

All sorted now. Sorry for the confusion and thanks for all the input.

 

[westward10]

So this had nothing to do with R1+R2 then

 

[shinypen]

Well yes but I didn't want to confuse things further, though I'd about used up all my help for one week.

I was taking a R1+R2 reading with the cpc still on the met and all mcbs open . I know this is wrong and can give multiple paths. Just a mistake. This was giving me a continuity reading (albeit a high one) when the light switch was open. I can't quite get my head around this. Was the high R1+R2 measured actually the path back through the sub-station??? Is that what you were saying Richard?

If Neutral and earth are connected in the CU bars and the main switch is closed then when the light switch is open the meter could measure a resistance loop, via the supply neutral earth connection, through the load and this could give a high reading of the sort of level you have measured.

I misread you Anthony and thought you were talking about continuity. Sorry.

No damp areas? Almost like there is something else in the circuit? Or when you IR'd did you have cpc on the bar still with neutral connected?

 

[Octopus]

The cpc continuity is now fine. I'd left cpc in cu met bar. When I removed it the readings were fine.

Hum .... I always test for IR with all the CPC's connected...

 

[Richard Burns]

I was taking a R1+R2 reading with the cpc still on the met and all mcbs open . I know this is wrong and can give multiple paths. Just a mistake. This was giving me a continuity reading (albeit a high one) when the light switch was open. I can't quite get my head around this. Was the high R1+R2 measured actually the path back through the sub-station??? Is that what you were saying Richard?

You can see from this diagram that if you were measuring continuity with the circuit fully connected but only the MCB open then the reading with the light switch closed would give a resistance close to the R1+R2 you are trying to measure, if you open the light switch then the only path left is through the load back to neutral and if the main switch is closed then the path can go via the neutral earth link back to the tester.
The high resistance of this path will only lower the R1+R2 resistance reading slightly when measured with the light switch closed (resistors in parallel).

 

[shinypen]

Thanks so much for posting this. Very helpful. The bit I couldn't get my head around is the main switch was open in my case. This would negate this path back?

I think I've tracked it down. It seemed to be a suspect pendant (it did have a small crack in it). In any case when I bypassed the pendant the rogue reading went away.

 

[Richard Burns]

Thanks so much for posting this. Very helpful. The bit I couldn't get my head around is the main switch was open in my case. This would negate this path back?

I think I've tracked it down. It seemed to be a suspect pendant (it did have a small crack in it). In any case when I bypassed the pendant the rogue reading went away.

In the case of the main switch being open then this would probably indicate a neutral earth fault on the circuit (or a switch which was not turning off effectively). The appearance of a resistance reading in the say 20 to 900 ohm range tends to indicate measurement through a resistive load, by whatever means. Though this is a broad generalisation.

 

[weevilward]

If you're not taking the neutral of the circuit you are testing out of the neutral bar at the CU, then you are effectively IR testing every circuit connected to that neutral bar. If any one of those circuits has a low IR then this will bring down your reading. Surge protectors plugged into a ring main will also show a low IR to earth. I think Powerbreakers show this too??? Also a combination of circuits with acceptable but low IR can combine to give an overall low IR that fails.

Always best to take both live conductors out at CU before IR testing. And I tend to test at 250v first to make sure there's nothing still plugged in or wired in before giving it full 500v. If L-N still shows continuity and I cannot find the thing wired into the circuit then I will just IR test st 250v with L and N tied, to avoid frying the thing. Chances are it's a TV amp or smoke detector or PIR or something equally sensitive to meggering.