Bonding water services when plastic pipes are present

[Rauer]

What are you actually doing in the property as There is a note at the front of each edition of BS7671, stating : "Existing installations that have been installed in accordance with earlier editions of the Regulations may not comply with this edition in every respect. This does not necessarily mean that they are unsafe for continued use or require upgrading." and then if you look at 132.16 it states that it must be adequate

 

[Engineer54]

I would isolate supply and remove main earth as for a Ze test. In turn removes all parallel paths from
Within the building, continuity test from this removed earth to the service.
Job done.

Why do you think that all parallel earth paths within a building always originate from the main incoming electrical earth??

 

[Deleted member 9648]

I would isolate supply and remove main earth as for a Ze test. In turn removes all parallel paths from
Within the building, continuity test from this removed earth to the service.
Job done.

What E54 says ^^^......could be a path from a gas service to the boiler....sructural steel etc,I always feel with this test that it's hard to be completly certain you are not being misled.

 

[rich.250]

I didn't mean they originate there. Just it removes all cpcs out of the question, I.e to the boiler,immersion etc where confusion can be found.

 

[malcolmsanford]

What E54 says ^^^......could be a path from a gas service to the boiler....sructural steel etc,I always feel with this test that it's hard to be completly certain you are not being misled.

When rich posted that he disconnected the Main earth when doing the 23Kohm test I asked why and understand why he did, but on thinking this over not sure if perhaps it is the right course.

We disconnect the main earth conductor for a Ze in order to prove an external earth path without parallel paths, but like a measured Zs I think that measuring a piece of metal work to prove it's an extraneous conductive part should be done with these paths in place. Because in my mind to prove this metal is an ECP should be done when the installation is "As-Is" not "How-Should" be.

I've argued this before, on here and off, when told a Zs should be calculated and not measured because of parallel paths, but my answer always is, You only get an electrical fault when an installation/circuit is LIVE, and I therefore when it is in use. So I want to know the condition and therefore the value of the Zs when the installation is in use. When a fault happens, the installation/circuit don't say hold on a second lads I need to disconnect all these parallel paths and make sure I can trip the OPD without them.

Yes I agree that perhaps 2 years down the line something can change within the installation, say removing a metallic service to a non metallic one, and those paths are removed, but I have never advocated designing an installation/circuit to rely on parallel paths, but it should be designed to trip the OPD without the paths. In my view these paths are literally a bonus.

So therefore the same with an ECP, I personally think by removing the main earth and those parallel paths your giving the installation a "false" state. The reason we do bond is to bring all these metallic services into the zone at the same potential. by removing your major source of earth aren't you removing your main reference point?

 

[johnboy6083]

The problem could be though malc: in a PME scenario, you could do the 23k ohm test with everything connected, and get a sufficiently low reading that you think its bonded. In fact, you could be relying on a 1mm conductor to carry diverted neutral currents.

I try to work on the basis that if I can find it I'll test it, if I can't find it then I'll pull a new bonding conductor in. I'm all for using structural steel ect though!

 

[yellowvanman]

I quizzed the NIC tech about it, he was saying i needed to use the adiabatic equation to work it out to see if it is sufficient enough, i've done this for main earth before
where the earth is sweated on the incoming cable but never for a services bond any ideas ?

I can see where the NIC guy is coming from. If you work out the main earth conductor size using the adiabatic, then the bonding conductor should be not less than half that and not less than 6mm. (544.1.1)

So if you get a minimum value of 11mm for your main earth (in which case 16mm should be installed) then min size of bonding would be 6mm as half 11mm is 5.5! This assumes that its not TNCS.

 

[oldtimer]

This is indeed a can of worms even if you disconnect all earths you can still get a low reading from the guy upstairs if you live in a flat ie his metal pipework could be exporting an earth to you through the pipework one other thing to note is that a property can lose an earth when they change say the gas/water supply pipes because they have no earth but they do earth bonding on the gas water hence they borrow an earth from next door this is becoming more common

 

[rich.250]

The problem could be though malc: in a PME scenario, you could do the 23k ohm test with everything connected, and get a sufficiently low reading that you think its bonded. In fact, you could be relying on a 1mm conductor to carry diverted neutral currents.

I try to work on the basis that if I can find it I'll test it, if I can't find it then I'll pull a new bonding conductor in. I'm all for using structural steel ect though!

Yes I agree, if you can't find the point of bonding I would re bond if under 23 k

 

[rich.250]

When rich posted that he disconnected the Main earth when doing the 23Kohm test I asked why and understand why he did, but on thinking this over not sure if perhaps it is the right course.

We disconnect the main earth conductor for a Ze in order to prove an external earth path without parallel paths, but like a measured Zs I think that measuring a piece of metal work to prove it's an extraneous conductive part should be done with these paths in place. Because in my mind to prove this metal is an ECP should be done when the installation is "As-Is" not "How-Should" be.

I've argued this before, on here and off, when told a Zs should be calculated and not measured because of parallel paths, but my answer always is, You only get an electrical fault when an installation/circuit is LIVE, and I therefore when it is in use. So I want to know the condition and therefore the value of the Zs when the installation is in use. When a fault happens, the installation/circuit don't say hold on a second lads I need to disconnect all these parallel paths and make sure I can trip the OPD without them.

Yes I agree that perhaps 2 years down the line something can change within the installation, say removing a metallic service to a non metallic one, and those paths are removed, but I have never advocated designing an installation/circuit to rely on parallel paths, but it should be designed to trip the OPD without the paths. In my view these paths are literally a bonus.

So therefore the same with an ECP, I personally think by removing the main earth and those parallel paths your giving the installation a "false" state. The reason we do bond is to bring all these metallic services into the zone at the same potential. by removing your major source of earth aren't you removing your main reference point?

I see what your satin malc, totally.
But isn't the point to see if it's extraneous.....?? Personally If I can't find the point of bonding I'd attach new.
But you could get a reading sufficiently low enough to make you think it is bonded!
Although now thinking about my own method, if I'm going to bond below 23k anyway I may aswell test with the ME in place...