2 ring mains connected to each other

[mikazza]

Hi,

I'm in the process of putting together plans to build an extension onto my house and i'm currently looking at the current electrical system thats installed. I'm no electrician and wont be doing the work myself i'm just trying to get a list of the work needing done together so I can get a price.

My main concern is that on the fuse box there are 2 circuit breakers for the sockets and 2 for the lights, meaning there are 2 circuits each for the lights and sockets, i've worked out the lights ok, one of the breakers controls the hallway and bedrooms and the other the rest of the house and have tested this by switching on lights and turning off the circuit breaker.

However on the sockets even though they have 2 x B32 breakers, you need to switch off both to cut the power, is this normal? or do I have to find out where these 2 circuits are joined and have it fixed.

I have attached a picture of the circuit breakers just incase it helps

 

[JohnW2]

However on the sockets even though they have 2 x B32 breakers, you need to switch off both to cut the power, is this normal? or do I have to find out where these 2 circuits are joined and have it fixed.

No, that is not normal, and is potentially dangerous. You therefore need to have it investigated and sorted out.

Kind Regards, John

 

[mikazza]

This is what I suspected, it's been like this for the past 10 years since I bought the house, i've been under the house trying to work out what wire went to what socket and they all look like they are joined up correctly, I suspect the issue the wiring in the circuit box itself.

 

[EFLImpudence]

I would suspect the four wires, two for each ring, have been mixed up and one from each ring has been connected to each MCB.

Nothing to do with the socket wiring.

 

[JohnW2]

I would suspect the four wires, two for each ring, have been mixed up and one from each ring has been connected to each MCB.

Indeed, that would seem to be the most likely explanation. I am probably not the only person who has (only once to my knowledge!) done that myself - although I did detect the error by testing prior to energising the circuits!

Kind Regards, John

 

[Taylortwocities]

As above. The problem is almost always in the consumer unit. This usually happens if the circuits have been disturbed at some time (eg a replacment consumer unit) and little or no testing was done afterwards.
It would be 15 minutes work for someone who knows what they are doing to sort it out.
Add it to the jobs for the electrician to do when he does the extension. Or you could use it as a qualifying task for your chosen sparky. Give him/her the job if he/she can sort it in less than 30 minutes!

 

[JohnW2]

As above. The problem is almost always in the consumer unit. This usually happens if the circuits have been disturbed at some time (eg a replacment consumer unit) and little or no testing was done afterwards.

Indeed - but, ironically, I have also seen it (**) arise as a result of testing - if the person doing the testing is silly enough to simultaneously disconnect the conductors for more than one of the circuits to be tested! I suppose that testing should also including ensuring that operation of just one MCB kills each circuit (after dead tests have been done) - but I don't think such a test is in 'the lists in the books', so probably doesn't often get done!

**It happened to a neighbour of mine fairly recently. He was certain that switching off just one MCB had previously killed one of his sockets circuits. However, immediately after he'd had an EICR undertaken, he asked for my advice because he had found that it was now necessary to switch off two MCBs to kill the circuit.

Kind Regards, John

 

[mikazza]

Thanks for all the help, i'm certain that will be where the problem is and will get it added onto my list, i'm not sure when this would have been put like that as no work has been done to the fuse box the entire time i've lived here, in the past before I bought the house garden, loft and garage circuit breakers have been added, and a whirlpool bath pump had been wired into a socket. Can't think of why any of that would require somebody to mess with the socket MCBs though, most likely the previous owners did it themselves.

 

[Risteard]

I don't think such a test is in 'the lists in the books', so probably doesn't often get done!

In the south of Ireland one of the required tests is for "erroneous connections" of phase conductors. Essentially each MCB/RCBO is switched off in turn and an insulation test is conducted between the open MCB and all of the other MCBs closed in parallel.

 

[JohnW2]

In the south of Ireland one of the required tests is for "erroneous connections" of phase conductors. Essentially each MCB/RCBO is switched off in turn and an insulation test is conducted between the open MCB and all of the other MCBs closed in parallel.

Interesting - but, as I said, I don't recall having seen such a test (or the simpler 'observational' one) in any of the 'lists' in (UK-orientated) books.

However, the test you describe sounds rather odd. Do you really mean that you connect the ('live') outputs of all the other MCBs together and test to them? If such a test were to be done, it would surely be better ('safer'?) to open all the MCBs, wouldn't it? Also, if there were any loads on any of the circuits, wouldn't connecting all these 'live' MCBs (probably protected by different RCDs) together probably result in RCD trips?

Kind Regards, John

 

[Risteard]

Do you really mean that you connect the ('live') outputs of all the other MCBs together and test to them? If such a test were to be done, it would surely be better ('safer'?) to open all the MCBs, wouldn't it?

If all MCBs were open then erroneous connections between phase conductors wouldn't be detected. Essentially you are probing the phase busbar with all circuits in parallel to the one you are proving has no erroneous connections to these. Obviously lamps/loads would need to be disconnected to avoid reading back through the neutral bar.

As for RCD trips an insulation resistance tester only utilises a current of between 1 and 2mA.

 

[JohnW2]

If all MCBs were open then erroneous connections between phase conductors wouldn't be detected. Essentially you are probing the phase busbar with all circuits in parallel to the one you are proving has no erroneous connections to these.

I don't really understand. Surely what one is interested in detecting are any 'erroneous connections' between the circuits connected to the 'outputs' of the MCBs - which, as I said, could be done with all MCBs open?

Obviously lamps/loads would need to be disconnected to avoid reading back through the neutral bar.

Indeed - and removing all loads (particularly 'hard wired' ones) from all circuits could be an pretty tedious job.

As for RCD trips an insulation resistance tester only utilises a current of between 1 and 2mA.

As I said, if one joined together the outputs of the (closed) MCBs of all but one of the circuits, if (as is likely) those circuits were protected by more than one RCD, and if there were any loads on any of the circuits (as above, potentially tedious to achieve), one or more RCDs could well trip even before you started your IR testing, couldn't they?

... or am I misunderstanding or missing something?

Kind Regards, John

 

[Risteard]

... or am I misunderstanding or missing something?

John, I fear you are missing something. This is a Pre-Connection (i.e. dead) Test.

I have no doubt that it is my fault for failing to explain this satisfactorily. If you bear with me I will attempt to find an explanation online (preferably with pictures) to make it clearer. (Unfortunately as there are far fewer ET101-centric resources than BS7671 resources that mightn't be too easy.)

Regards,
Risteard

 

[Risteard]

Have a look here: http://www.airm.ie/system/download_...cal safety - RECI presentation.pdf?1358438070

Page 44 is the one you want.

 

[JohnW2]

John, I fear you are missing something. This is a Pre-Connection (i.e. dead) Test.

Ah, fair enough - as you will presumably have worked out from everything I was saying, I did not realise that you were talking about a totally dead test (i.e. with the L busbar not live).

It also now makes sense in that closing all of the 'other' MCBs is simply a means of connecting the associated circuits all in parallel - I had envisaged that you were talking about far more tedious was of achieving that!

No further explanation is necessary.

Kind Regards, John