Mains power cable on surface

[DaveOC]

Hi, My meter is outside my front door, the cables go through the house wall and the run up the inside wall to the consumer unit which is above the front door, the cables are in plastic trunking. I would like sink the cables into the wall so they are not seen. Is this allowed and can an electrician do it or will the power provider have to do it?
Thanks

 

[Chunky19]

The cables between the meter and your consumer unit are your property. The only problem is isolating them to safely do the job. Speak with your electrician and he can see it and advise.

 

[davelx]

Cables sunk into wall without mechanical protection have to be RCD protected, but you can't RCD protect meter tails - so they would have to be in conduit or steel channelling, I think.

 

[AndyPRK]

Sadly it’s very common for tails to be inside a wall and unprotected

 

[davelx]

Sadly it’s very common for tails to be inside a wall and unprotected

I'm sure it is - but we shouldn't encourage it!

 

[plugwash]

Often the tails are run in the cavity, assuming a brick/block inner skin this puts them deep enough to avoid the RCD protection reg, but there may be concerns about thermal insulation or cavity bridging.

 

[JohnW2]

Cables sunk into wall without mechanical protection have to be RCD protected, but you can't RCD protect meter tails - so they would have to be in conduit or steel channelling, I think.

I don't think there is any theoretical reason why an RCD could not be installed in, say, the external meter cabinet, is there, and I see nothing in 522.6.202 which precludes the use of a time-delayed one (thereby addressing the possibility of 'whole installation power loss' in the event of a fault).

However, give that the CU is 'above a door', I think that one might struggle to get the tails all within 'safe' zones, which would be the other requirement in the absence of mechanical protection

Kind Regards, John

 

[plugwash]

My understanding is if you want discrimination between RCDs you would want both a "time delay" on the first RCD and a difference in trip current.

So even if you could find a 30mA time delayed RCD (does anyone make them?) and even if it complied with the regulations for concealed cables (I'm no expert on that) you would then have to use 10mA RCDs on the individual circuits.

In other words not very practical.

 

[JohnW2]

My understanding is if you want discrimination between RCDs you would want both a "time delay" on the first RCD and a difference in trip current.

I don't see why that should be the case. If the downstream (not delayed) one trips long before the delay period of the upstream one, then the latter wouldn't trip, even if it had the same IΔn ('trip current'), would it? However ....

So even if you could find a 30mA time delayed RCD (does anyone make them?) ...

... I had overlooked the fact that, as you imply, time-delayed 30mA RCDs (which would be required for protection of buried cables) probably do not exist - in which case the question would become moot.

In any event, as I said, I thing it would have been a struggle (impossible) to get buried tails to a CU 'above a door' with the cables being entirely within 'safe' zones - so perhaps another reason why the question is moot!

Kind Regards, John

 

[davelx]

Never mind.. please delete!

 

[EFLImpudence]

The time delay only matters if both or all RCDs would trip with a particular fault.

Ignoring the allowed tolerances; if you have a 10mA leakage then only a 10mA RCD will trip; with a 30mA leakage a 10mA and 30mA will trip; a 300mA leakage a 10mA, 30mA and 300mA will trip - not the 300mA if TD.

As you know, 30mA RCDs are required in situations for personal protection so, presumably, time delayed ones are never required.

 

[plugwash]

I don't see why that should be the case. If the downstream (not delayed) one trips long before the delay period of the upstream one, then the latter wouldn't trip, even if it had the same IΔn ('trip current'), would it? However ....

Trip points of protective devices aren't an exact science. My understanding is that a "30mA" RCD may have an actual "trip current" anywhere between 15mA and 30mA.

So the time delay provides discrimination in the case of sudden faults while the difference in nominal trip current provides discrimination in the event of gradual faults.

 

[AndyPRK]

I would say it’s common to have large leakage currents under fault conditions.
(In excess of 500ma)
I’m thinking of a failed heating element

 

[plugwash]

And I'm thinking of a fitting where the seals aren't quite right letting in gradually more moisture, or a capacitor failing gradually.

So I think you need both a timing difference (for the fast faults) and a setpoint difference (for the slow faults) to achive effective discrimination.

 

[JohnW2]

Trip points of protective devices aren't an exact science. My understanding is that a "30mA" RCD may have an actual "trip current" anywhere between 15mA and 30mA.

It gets a bit more messy with lower residual currents, but with a residual current of 5 x IΔn (e.g. 500 mA for a 100mA RCD), the maximum trip time for a non-delayed RCD is 40 ms and the minimum trip time for a delayed one is also 40 ms - so (unless both tripped in exactly 40 ms), the non-delayed one would always trip before the delayed one's minimum trip time.

However, as I said, it's really all a bit moot, not the least because the device that would be required probably doesn't esist!

Kind Regards, John