Outdoor lights on two ring circuits

[paul798]

Hello,

We've had an extension built a few years ago and are finally getting round to install some outdoor lights. The electrician at the time left the cables in place for us to connect the lights when ready without giving any further instructions. The plan was always to install two outdoor lights, one on either side of the extension and both operated via a light switch on the inside. When turning off the circuit breaker for the lighting circuit at the consumer unit today I noticed that only cable 1 stopped being live while 2 and 3 remained live. Upon investigation I found that only cable 1 responds to the light switch inside the house. Cables 2 & 3 remain live regardless of the position of the light switch and they also remain live if either the circuit breaker for the lighting or socket ring remain on. I have to turn off the circuit breakers for both rings before cables 2 & 3 stop being live. I've tested this with a LAP non-contact voltage detector pen. I know it is not exactly the gold standard but has worked reliably elsewhere in the house.

Normally I would get the electrician / builder back to explain this but given it has been a few years that is not really an option. Is this setup ok and if so, how should the cables be connected for the lights? Up until today I simply assumed cable 2-3 were an extension.

 

[oceansoul]

How are you testing these are live?

 

[EFLImpudence]

I've tested this with a LAP non-contact voltage detector pen. I know it is not exactly the gold standard but has worked reliably elsewhere in the house.

 

[ericmark]

they also remain live if either the circuit breaker for the lighting or socket ring remain on.

That's the worrying bit, and may have nothing to do with outside lights. Do other items become dead if only one overload turned off?

Personally I never turn the outside lights off at the wall switch, I use smart bulbs and only turn off for changing bulbs, so no real problem not turning off, it is the possibly that some thing is connecting the sockets and lights together.

Nothing wrong with a non contact voltage tester I use one myself but there are varying quality of tester, mine rather cheap at £35 but seems to do a good job. But it references to the leads plugged in or your body which act as aerials, so if your live then the tester will pick up ground instead, it can work either way around, and this is the problem with all non contact voltage testers.

 

[oldbutnotdead]

First thing- well done for actually testing before charging in.
Next step is to use a real meter. You may find that cable 2 and 3 are 2 ends of 1 bit of cable. Since they are connected to nothing they'll be picking up a bit of voltage by induction.
Isolate supply to house (just to make sure).
Strip ends of all 3 cables, put terminal block or wagos on each end
Power on.
Use your real meter to check line 1 has 240v ish live to neutral and live to earth.
With one probe on the earth you've proved (on line 1), check all 3 cores on line 2 for voltage. Ditto for line 3 (may have to extend your earth probe).
Let us know what you find

EDIT Do these tests with the mcbs you mentioned earlier in both positions.

 

[paul798]

Thank you so much! Should have used a multimeter from the start.

* Line 1 had 240V live to neutral and live to earth
* Line 2 had 0V on all cores when either or both of the mcbs were off. When they were both on I could see between 0.2V and 0.8V whenever I tested the cores, in all cases this dropped to 0V after a second
* Line 3 I couldn't test as it is about 7m away from line 1 & 2

Given the intended setup I think your explanation that this was simply a case of induction confusing the tester pen sounds most likely. Thank you again!

 

[Taylortwocities]

Those pens light up if you rub them on yer jumper!

 

[oldbutnotdead]

Thank you so much! Should have used a multimeter from the start.

* Line 1 had 240V live to neutral and live to earth
* Line 2 had 0V on all cores when either or both of the mcbs were off. When they were both on I could see between 0.2V and 0.8V whenever I tested the cores, in all cases this dropped to 0V after a second
* Line 3 I couldn't test as it is about 7m away from line 1 & 2

Given the intended setup I think your explanation that this was simply a case of induction confusing the tester pen sounds most likely. Thank you again!

You can use an extension cable to give a a proved earth near cable 3....
Once you know there are no real volts on it it'll be worth checking it really is the other end of cable 2....

 

[bernardgreen]

I would suggest

Cable 1 is from the switch bring Switched Live and Neutral to the first lamp

Cable 2 and Cable 3 are the two ends of the same cable which will take Switch Live from the first lamp the second lamp.

To confirm this without the nead for a long wander lead

With the power OFF test for any voltage between Live and Neutral on both Cable 2 and Cable 3

If there is no voltage then connect Live and Neutral of cable 2 together in a terminal block
Test for continuity between Live and Neutral in Cable 3 (test A)

Remove the connection between Live and Neutral at Cable 2
Test for continuity between Live and Neutral in Cable 3 (test B)

If (test A) shows continuity and (test B) shows open circuit then it is almost 100% certain that Cable 2 and Cable 3 are two ends of the same cable.

 

[paul798]

You can use an extension cable to give a a proved earth near cable 3....
Once you know there are no real volts on it it'll be worth checking it really is the other end of cable 2....

I used the extension and found no voltage on cable 3 when connected to the known good earth. I then went ahead and connected cable 1 & 2 which resulted in 240V on cable 3 that could be turned on/off via the light switch. At that point I was happy, installed all the lights and it has been working just fine.

Thanks everyone for your help! Next time I'll make sure to use a multimeter from the start.

 

[Harry Bloomfield]

Thanks everyone for your help! Next time I'll make sure to use a multimeter from the start.

Every day is a school day!