Testing extraneous conductive part

[Zdb]

Just been read through this and it says:

How to determine if it’s an extraneous-conductive-part

For this, a simple measurement is undertaken. This is done by testing the resistance between the suspected extraneous-conductive-part and the MET or the nearest known connection to Earth, such as the circuit protective conductor (CPC) of a local circuit (so long as its connection to Earth has been verified).

Published guidance by the IET suggests that if the ohmic value between the suspected extraneous part and Earth is greater than 22 kΩ, then the part in question is considered as unable to introduce a shock risk and, as such, won’t require bonding.

To measure this sort of reading will require the use of the insulation resistance testing range (MΩ) where 22 kΩ will usually be displayed as 0.02 MΩ.

If the test reading indicates anything greater than 0.03 MΩ, then it may be concluded that the item in question isn’t likely to introduce a dangerous potential and therefore wouldn’t require bonding.

So does this mean that using a MFT set to insulation resistance, I can connect one lead to some cable basket and the other to a nearby socket using a plug (R2) adapter instead of the MET?

 

[telectrix]

So does this mean that using a MFT set to insulation resistance, I can connect one lead to some cable basket and the other to a nearby socket using a plug (R2) adapter instead of the MET?


yes as longas you have verified that the socket earth is connected back to earth . this may be done with a simple R1 + R2.

 

[Ian1981]

Just been read through this and it says:

How to determine if it’s an extraneous-conductive-part

For this, a simple measurement is undertaken. This is done by testing the resistance between the suspected extraneous-conductive-part and the MET or the nearest known connection to Earth, such as the circuit protective conductor (CPC) of a local circuit (so long as its connection to Earth has been verified).

Published guidance by the IET suggests that if the ohmic value between the suspected extraneous part and Earth is greater than 22 kΩ, then the part in question is considered as unable to introduce a shock risk and, as such, won’t require bonding.

To measure this sort of reading will require the use of the insulation resistance testing range (MΩ) where 22 kΩ will usually be displayed as 0.02 MΩ.

If the test reading indicates anything greater than 0.03 MΩ, then it may be concluded that the item in question isn’t likely to introduce a dangerous potential and therefore wouldn’t require bonding.

So does this mean that using a MFT set to insulation resistance, I can connect one lead to some cable basket and the other to a nearby socket using a plug (R2) adapter instead of the MET?

No not really as all parallel paths would need to be removed. The test should be from the means of earthing or connection to the earthing conductor with all possible other connections to earth via different routes removed

 

[cliffed]

No not really as all parallel paths would need to be removed. The test should be from the means of earthing or connection to the earthing conductor with all possible other connections to earth via different routes removed

Agree,near impossible in an energised installation with the parallel earths,& you really wouldn’t want to disconnect All those earths would you.
So if you think it needs Main bonding then ,Bond it.

 

[Midwest]

Surely removing those parallel earth paths would for when testing Ze, for example.

In this case testing is to see whether a piece of metal is an extraneous conductive part, or is just a piece of metal. Those potential parallel earth paths, may have a deciding factor on whether it needs bounding or not.

The test procedure, as I've known it, is that highlighted in red by @Zdb. Otherwise, when the parallel earths, such as circuit CPC's are reconnected, that might alter the value obtained, and the requirement or not to bond.

 

[radiohead]

This test is often almost impossible in practice. A copper pipe for example is quite likely to be mechanically connected to another extraneous part such as a gas service (via the boiler)or structural steel (from fixings). In reality the test is only conclusive where it can be visually ascertained that the metallic part is completely independent of other possible extraneous parts.

 

[Ian1981]

Surely removing those parallel earth paths would for when testing Ze, for example.

In this case testing is to see whether a piece of metal is an extraneous conductive part, or is just a piece of metal. Those potential parallel earth paths, may have a deciding factor on whether it needs bounding or not.

The test procedure, as I've known it, is that highlighted in red by @Zdb. Otherwise, when the parallel earths, such as circuit CPC's are reconnected, that might alter the value obtained, and the requirement or not to bond.

It requires parallel paths removed,for example the cpc connected to the boiler earthing which in turn is in contact with the copper gas pipe. If you do not disconnect the cpc if testing that the gas pipe requires bonding or it’s just an isolated piece of metal, then it will measure practically a dead short 0.00 Mohm on the instrument

 

[Midwest]

It requires parallel paths removed,for example the cpc connected to the boiler earthing which in turn is in contact with the copper gas pipe. If you do not disconnect the cpc if testing that the gas pipe requires bonding or it’s just an isolated piece of metal, then it will measure practically a dead short 0.00 Mohm on the instrument

But that kinda defeats the object of the test. If it’s a dead short, because of its physical connection with the boiler, then cpc. What’s the point of deciding to bond, if the test result suggest it should be.

On the other hand, what if the physical connection is not great, and would give a value over 1667 ohms, but less than 0.02 MO. With the CPC disconnected, it then gives a value of over 0.02 MΩ.

I haven’t my Guidance notes to hand, but all references I can google for this test, are as said before; between extraneous part and MET or known earth, such as CPC. No mention of dis’ parallel earths. Does it say something different in the Guidance notes?

 

[Ian1981]

But that kinda defeats the object of the test. If it’s a dead short, because of its physical connection with the boiler, then cpc. What’s the point of deciding to bond, if the test result suggest it should be.

On the other hand, what if the physical connection is not great, and would give a value over 1667 ohms, but less than 0.02 MO. With the CPC disconnected, it then gives a value of over 0.02 MΩ.

I haven’t my Guidance notes to hand, but all references I can google for this test, are as said before; between extraneous part and MET or known earth, such as CPC. No mention of dis’ parallel earths. Does it say something different in the Guidance notes?

The test can’t be made where a conductive part say for example a conductive part like steel is to be tested but you have electrical items and say metallic containment like conduit connected to the steel by ways of fixings, bolts etc.
it would be impossible to test it as it’s in direct contact with exposed conductive parts that are connected together at points within the installation, so a test between a known cpc and said steel would be pointless unless you can remove the parallel paths created by the cpc’s or exposed conductive parts as you would essentially read 0.00 Mohms when trying to test.
same as a gas pipe connected to the boiler manifold / metallic housing, along with the cpc from the heating circuit, they are connected together so this connection must be removed.
By way of circumstance you always end up having metallic parts connected to earth that don’t need to be but bonding is entirely different to earthing.

 

[telectrix]

The test can’t be made where a conductive part say for example a conductive part like steel is to be tested but you have electrical items and say metallic containment like conduit connected to the steel by ways of fixings, bolts etc.
it would be impossible to test it as it’s in direct contact with exposed conductive parts that are connected together at points within the installation, so a test between a known cpc and said steel would be pointless unless you can remove the parallel paths created by the cpc’s or exposed conductive parts as you would essentially read 0.00 Mohms when trying to test.
same as a gas pipe connected to the boiler manifold / metallic housing, along with the cpc from the heating circuit, they are connected together so this connection must be removed.

but in those cases, the "suspect" metal is already bonded due to it being bolted/welded/etc. to the bonded extraneous parts.

 

[Ian1981]

but in those cases, the "suspect" metal is already bonded due to it being bolted/welded/etc. to the bonded extraneous parts.

No it maybe earthed by being in contact with exposed parts but not necessarily bonded which is different

 

[telectrix]

No it maybe earthed by being in contact with exposed parts but not necessarily bonded which is different

earthed if connected to exposed conductive parts, but bonded if connected to already bonded extraneous parts.

 

[Ian1981]

earthed if connected to exposed conductive parts, but bonded if connected to already bonded extraneous parts.

Something like that ha
I’ve always gone down the route of testing it with the installation isolated, earthing conductor disconnected from the earthing terminal in the CU but connected to the means of earthing,then from the disconnected earthing conductor to the suspect metal.
In practice tho I’ve only ever done this at a domestic level, a commercial or industrial installation already in service, I find is impossible to do the test.

 

[Midwest]

Here's something I've read, they might be wrong of course.
SO HOW DO WE MEASURE IT?

And Does Metallic Cable Tray Require Bonding? - https://professional-electrician.com/technical/bonding-time/

 

[ipf]

earthed if connected to exposed conductive parts, but bonded if connected to already bonded extraneous parts.

Like in a factory, say, where you have the bonded main steel structure and take off a bond for any other required...

 

[Ian1981]

Here's something I've read, they might be wrong of course.
SO HOW DO WE MEASURE IT?

And Does Metallic Cable Tray Require Bonding? - https://professional-electrician.com/technical/bonding-time/

Yeah i would that’s fair enough as long as you haven’t earthed the tray by ways of fixing exposed conductive parts to it or say fixed the tray to steel which has metallic conduit fixed to it also, this would then earth all metallic parts concerned, so it wouldn’t be easy to test to see if it was extraneous ( which I can’t see how it would be in most circumstances)

 

[telectrix]

Like in a factory, say, where you have the bonded main steel structure and take off a bond for any other required...

Egggsackly. free range. ( 3 friggin' words)

 

[Midwest]

Yeah i would that’s fair enough as long as you haven’t earthed the tray by ways of fixing exposed conductive parts to it or say fixed the tray to steel which has metallic conduit fixed to it also, this would then earth all metallic parts concerned, so it wouldn’t be easy to test to see if it was extraneous ( which I can’t see how it would be in most circumstances)

And the other one?

 

[Ian1981]

And the other one?

Same principle, as long as you haven’t earthed it.

 

[Midwest]

Hmmm, I think your wrong about this concept. The principal of the test is simple; no one mentions disconnecting anything.

 

[Ian1981]

But what is the use of testing between 2 points if they are connected together somewhere in the installation?
Like for instance a gas pipe and a known earth, but the gas pipe is connected at one end to the boiler which is connected to earth by a cpc. The cpc needs disconnecting at the boiler end then test between earth and the pipe.
That’s just one example you do not necessarily have that scenario every time of having to disconnecting anything.

 

[Midwest]

Thats the idea of this test; to discover if somehow two points are connected in someway or not.

 

[telectrix]

you could always poke it with a "live" wire. if it goes bang, it's extraneous. .....trainees.DON'T TRY THIS AT HOME>

 

[Ian1981]

Thats the idea of this test; to discover if somehow two points are connected in someway or not.

Extract taken from NAPITS site guide, the point is actually to find out if the metallic conductive part is introducing an earth potential into the installation and by testing between the earth of the installation and the conductive part, that the conductive part under test has a high enough resistance to earth that it doesn’t require bonding to the earthing arrangement of the installation

It’s

 

[static zap]

Inverse logic -Bang once then no more ? Now cured.

if it goes bang, it's extraneous.

---NOT good advice -- but get's you thinking !
22k more of a fizzle ..
(Door bell /out door LV lighting /CCTVprobably broke now)

 

[Midwest]

Extract taken from NAPITS site guide, the point is actually to find out if the metallic conductive part is introducing an earth potential into the installation and by testing between the earth of the installation and the conductive part, that the conductive part under test has a high enough resistance to earth that it doesn’t require bonding to the earthing arrangement of the installation

Fair enough. But that’s just one piece of guidance that says dis the cpc’s, against the two I posted, which suggested using cpc’s as the known earth.

Does Guidance Note 3 or 8 make any suggestions?

 

[Ian1981]

Fair enough. But that’s just one piece of guidance that says dis the cpc’s, against the two I posted, which suggested using cpc’s as the known earth.

Does Guidance Note 3 or 8 make any suggestions?

It’s only relevant to disconnect the cpc’s if required, you don’t always need to if the metallic part has not been connected together with a cpc to earth .
I don’t think gn3 covers this subject but gn5 and 8 do so I’ll have a dig out

 

[Lister1987]

you could always poke it with a "live" wire. if it goes bang, it's extraneous. .....trainees.DON'T TRY THIS AT HOME>

But I'm ok to do it in colleege....right? ?

 

[Zdb]

Thanks for all of the replies.

Interesting topic. Might have to invest in a copy of GN8.

 

[Midwest]

Thanks for all of the replies.

Interesting topic. Might have to invest in a copy of GN8.

You can tell us the answer then

 

[Zdb]

It says in GN8 to test between the MET and the suspected extraneous conductive part.

Not a problem in a domestic installation but a known/proven connection to earth is more practical in a commercial/industrial scenario.

 

[Zdb]

So, I carried out this test yesterday on some cable basket I put up for some Cat6 data cables. I tested between a known earth from a socket outlet and the cable basket and I got a reading of 0.47 ohms. Nice low reading via parallel paths (earthed structural metal work). I then bonded the basket using some 6mm.

So am I right in thinking that this needed bonding due to the reading, and if the reading had been above 23,000 ohms then it wouldn't have needed bonding?

 

[Midwest]

I don't think it needed bonding. Here's a view from someone else;

Does a Metallic Cable Tray Require Earthing or Bonding? | Tradeskills4u

A. In short the answer is ‘depends’ but the answer could be ‘earthing’, ‘bonding’ or ‘neither’ depending on the circumstances. Looking at each of these in turn: If the tray is an exposed conductive part it requires earthing. If the tray is and extraneous conductive part it requires bonding. If...

www.tradeskills4u.co.uk

 

[Mike Johnson]

It's usual in France to have one of these between the earth rod and the DB on the outside wall of any domestic property: Barrette de coupure de terre haute P.65 mm LEGRAND | Leroy Merlin - https://www.leroymerlin.fr/v3/p/produits/barrette-de-coupure-de-terre-haute-p-65-mm-legrand-e1401450423 makes checking extraneous conductive parts relatively easy loosen the top nut and swing the brass joiner out of the way.

 

[static zap]

"Barrette de coupure de terre haute"

makes checking extraneous conductive parts relatively easy loosen the top nut and swing the brass joiner out of the way.

May be clamp it first ?

 

[Mike Johnson]

Clamp what where?

 

[static zap]

Clamp what where?

Make sure no current is entering / leaving that exotic french "link"
before opening it and having it turn into something that may unexpectedly exceed "touch voltage limits" !
(resp to your sig---- all my fingers too !)

 

[Mike Johnson]

If any current was entering that exotic French link then the Disconjoncteur tetrapolair de branchement would trip, can't see how any current could enter the TT earthing system without tripping something, perhaps even the Parafonudre Disconjoncteur Quata would go first.

 

[static zap]

Are you sure ?
Is

the Disconjoncteur tetrapolair de branchement would trip,

All equipment 100% reliable , trustworthy !
Could be staking your / someone else's life on the assumption?

 

[Mike Johnson]

There are safety procedures that we all go through before we carry out any work on a system, why are you persevering with this premise?

 

[i=p/u]

Is tray part if the electrical system or not, yes it is earth it.