Help tracing ring main in singles

[Alan Marriott]

Hi
I’m after some advice please. I’m on a job where there’s a junction box with ring main wired in singles through 50x50 galv trunking. I need to break into this ring for my additional sockets but need to identify which of the cables are incoming and outgoing ones!? I can’t follow cable in trunking as it’s butt up against roof and can’t get my fingers in and cables also tangled round others in there. What’s best method to find out which cables are incoming and outgoing circuits?

Thanks

 

[i=p/u]

Bell them out if in doubt , sometimes you might be able to see tape from whoever pulled them in

 

[Julie.]

If everything can be made dead, if you can inject a decent current in one end and back (low voltage transformer like the train set ones with an adjustment), then you can use a current clamp meter to identify the cores carrying current.

Lot of a faff - usually easiest to pull cables back and forth with an able assistant

 

[timbobelfast]

I would try plugging in a known load such as a heater and use a clamp meter to trace the correct cores. It's not scientific, but its a method I've used before when tracing single core neutrals in a DB.

 

[timhoward]

Normally I'd identify a set of 3 at the CU, break the ring (safely) at furthest point, connect a load, only connect up one set of 3 at CU, and use a clamp meter to identify the L and N showing current.
But I did note that you said you can't get fingers in, so probably no room for a clamp meter either. It sounds like inserting joints is going to be very difficult too even if you find the right cables.

Can you instead disconnect at the next nearest socket, one colour at a time, hook and tape new cables to the ones you disconnected, and pull each in turn at the joint box until one moves, and pull a new length through - then you'd also have some slack to work with for your new connections?

 

[Lucien Nunes]

There's umpteen ways of doing it with an MFT or multimeter. Probably the best as it doesn't involve opening any other fittings:
1. Isolate circuit, separate conductors at JB, connect one L to one N, leave the other two floating. Test resistance at a number of sockets. If the readings vary progressively the joined L & N are from the same leg of the ring, if they are all similar, the L & N are from opposite legs of the ring.

 

[Lister1987]

There's umpteen ways of doing it with an MFT or multimeter. Probably the best as it doesn't involve opening any other fittings:
1. Isolate circuit, separate conductors at JB, connect one L to one N, leave the other two floating. Test resistance at a number of sockets. If the readings vary progressively the joined L & N are from the same leg of the ring, if they are all similar, the L & N are from opposite legs of the ring.

I would welcome an article on the different methods of this sort of thing, same with belling out, I know it's a 'simple' thing but it's one I struggle to get my head around ?

 

[i=p/u]

Lister touch both ears with bell tester for example.?

 

[timhoward]

I would welcome an article on the different methods of this sort of thing, same with belling out, I know it's a 'simple' thing but it's one I struggle to get my head around ?

A variation on the method above is useful for finding out the route of a ring circuit. When ring is not continuous I do this. Both ends out at CU and join two conductors on the same leg one end. Take resistance readings at each socket. If ring is complete then ring circuit route is highest to lowest, lowest being closest to the joined end, highest being closest to the open end. (Spurs can slightly confuse this)
If there’s a break then the last socket with a reading and the first without is where to start.
If IR tests fail I sometimes do this to find the middle to decide where to start breaking it down.