IET
Decrease font size
Increase font size
Topic Title: Determining an extraneous conductive-part by measurement
Topic Summary:
Created On: 12 April 2018 11:26 PM
Status: Read Only
Linear : Threading : Single : Branch
<< 1 2 3 4 5 6 Previous Next Last unread
Search Topic Search Topic
Topic Tools Topic Tools
View similar topics View similar topics
View topic in raw text format. Print this topic.
 17 April 2018 11:45 AM
User is offline View Users Profile Print this message



AJJewsbury

Posts: 17079
Joined: 13 August 2003

I think that is fundamentally what I have been saying, isn't it?

Hopefully it's what you've both been saying, if using differing words!

However, where RCDs are fitted, isn't the Ia, the I delta n of the RCD making the value 1,666 Ohms instead of 0.3125?

If there are 30mA RCD protection on all circuits of the entire installation - then yes I'd agree. If however there are any distribution circuits or final circuits without 30mA RCD protection (still common on existing installations of course, even those with refurbished bathrooms) then much lower R values may need to be considered. RCDs on bathroom circuits limit the duration of faults within the bathroom, but can't limit the duration of faults imported from elsewhere in the installation via extraneous-conductive-parts.

- Andy.
 17 April 2018 05:04 PM
User is offline View Users Profile Print this message



Alcomax

Posts: 334
Joined: 12 November 2009

If however there are any distribution circuits or final circuits without 30mA RCD protection (still common on existing installations of course, even those with refurbished bathrooms) then much lower R values may need to be considered


Perhaps lower than we would expect if we were not relying on a final circuit OPD without RCD.

There is of course the TN system and metal consumer unit and so the supply fuse presents the worst case scenario, much lower than 0.3125 ohms.

Pre- 16th in existing installs with no, or difficult to achieve neatly, supplementary bonding, would tie myself in knots deciding what resistance value between metal bits in bathroom was "safe"....then would generally default to it is safe enough if main bonding was in place and there was a 30mA RCD upfront of all circuits and / or before metal DB , if there were one.

The value 0.05 ohms was / is still guidance to confirm supplementary bonding is "effective". That figure was I believe an arbritary value based on meter accuracy, but also seems to be in the region of what is required for a 100 amp supply fuse.
 17 April 2018 05:22 PM
User is offline View Users Profile Print this message



AJJewsbury

Posts: 17079
Joined: 13 August 2003

There is of course the TN system and metal consumer unit

I did think about that case, but decided not to mention it! My thinking was that a fault at or very near to the MET would raise the potential of the installation's entire protective system pretty much equally (rather like a fault outside the installation) - so of itself shouldn't cause a significant p.d. within the bathroom. More of a worry, I think, are faults on long circuits with relatively large R2 vales - creating a higher voltage difference between the fault and the MET. But certainly a good long submain on a relatively large fuse or MCB could be even worse than my example and require significantly lower values.

- Andy.
 17 April 2018 05:34 PM
User is offline View Users Profile Print this message



Weirdbeard2

Posts: 496
Joined: 29 November 2017

Originally posted by: gkenyon

Originally posted by: geoffsd



Originally posted by: gkenyon







If the CH pipes are metal, and they are an "extraneous-conductive-part" for the location (bathroom), then EITHER:







(a) Main bond them; OR







(b) Provide supplementary local protective bonding in the bathroom.




No, neither is necessary - if all the conditions of 701.415.2 are met.
But they can't be met if the CH pipes are extraneous-conductive-parts of the bathroom (even if they are not for the installation). Even if the pipes are NOT extraneous-conductive-parts of the installation, this can occur if the pipes are earthed, say by the protective earthing of a boiler which is located away from the bathroom.





Hi GK, if the pipes are earthed by a protective conductor they are exposed conductive parts as they will become live in the event of a fault, they cannot be both extraneous and exposed conductive parts at the same time?
 17 April 2018 05:51 PM
User is offline View Users Profile Print this message



AJJewsbury

Posts: 17079
Joined: 13 August 2003

Hi GK, if the pipes are earthed by a protective conductor they are exposed conductive parts as they will become live in the event of a fault, they cannot be both extraneous and exposed conductive parts at the same time?

The pipes can't be exposed-conductive-parts as they're not part of electrical equipment (see the part 2 definitions) - where they can however introduce a potential due being in contact with an exposed-conductive-part (or true earth) they do qualify as extraneous-conductive-parts. (I would agree that the definitions in this area are far from clear though - especially where 'second hand' fault voltages are concerned.)

- Andy.
 17 April 2018 06:10 PM
User is offline View Users Profile Print this message



Zoomup

Posts: 4923
Joined: 20 February 2014

I was working in a 1989 built first floor flat today. It is the highest flat. The bathroom had been re-tiled and all pipes were boxed in. No pipes are visible or accessible. The airing cupboard is sited away from the bathroom in a back bedroom. The only metal water features in the bathroom were a basin mixer tap and a shower head over the bath. No pipes could be seen at all. No confirmation of bonding could be seen. The main water pipe into the flat could not be seen anywhere. Any water pipes in the kitchen were hidden by units and panelling. A 10mm2 green and yellow cable could be seen for a run of about 2 meters in the loft running towards the kitchen, but terminated where? I presume that the incoming water pipe is plastic, but can not confirm this as it can not be seen nor any main bonding earth clamp. The metal sink top is earthed but this could be via the immersion heater. The loft has deep, deep glass fibre insulation so may cover up anything. So it is very difficult to draw any conclusions about the main and supplementary bonding.

What do others think?

Z.
 17 April 2018 06:47 PM
User is offline View Users Profile Print this message



geoffsd

Posts: 2110
Joined: 15 June 2010

Originally posted by: Alcomax

The value 0.05 ohms was / is still guidance to confirm supplementary bonding is "effective". That figure was I believe an arbritary value based on meter accuracy, but also seems to be in the region of what is required for a 100 amp supply fuse.

0.05 Ohms is a value accepted as near negligible and is used to indicate a good connection between the conductor and the bonded part.

It is not a relevant value between parts - although, obviously, if only a short distance it will be the case.

The effectve value between parts is, as above, 50/Ia or 50/I delta n.
 17 April 2018 07:35 PM
User is offline View Users Profile Print this message



Weirdbeard2

Posts: 496
Joined: 29 November 2017

Originally posted by: AJJewsbury

Hi GK, if the pipes are earthed by a protective conductor they are exposed conductive parts as they will become live in the event of a fault, they cannot be both extraneous and exposed conductive parts at the same time?


The pipes can't be exposed-conductive-parts as they're not part of electrical equipment (see the part 2 definitions)



- Andy.


Andy, thanks for the reply, but pipes are allowed as earth electrodes let alone protective conductors, which seems to me that they generally qualify as electical equipment especially where a fault will cause them to become live?
 19 April 2018 07:10 PM
User is offline View Users Profile Print this message



Weirdbeard2

Posts: 496
Joined: 29 November 2017

Originally posted by: MichaelGranvilleLevens




An NICEIC assessor advised me not only the main water inlet which was already main bonded with 10mm but also the central heating pipes all needed to be main bonded too as they go into the ground, which I have now done.


He advised that I would need a test reading of 50,000ohms ( basically the non conducting location regs) in order for it not to be required.


Would it be fair to summarise that the assessor is incorrect?
 19 April 2018 08:06 PM
User is offline View Users Profile Print this message



geoffsd

Posts: 2110
Joined: 15 June 2010

No.



Originally posted by: Weirdbeard2

but pipes are allowed as earth electrodes

If you mean water pipes; not any more.

which seems to me that they generally qualify as electical equipment especially where a fault will cause them to become live?

I think 'equipment' means appliances, fittings and the like.
If you consider a pipe connected to, for example, a shower is likely to become live by contact then it would be an exposed-c-p and should be earthed.
 19 April 2018 09:00 PM
User is offline View Users Profile Print this message



Weirdbeard2

Posts: 496
Joined: 29 November 2017

Originally posted by: geoffsd

No.


Originally posted by: Weirdbeard2

but pipes are allowed as earth electrodes


If you mean water pipes; not any more.

which seems to me that they generally qualify as electical equipment especially where a fault will cause them to become live?


I think 'equipment' means appliances, fittings and the like.

If you consider a pipe connected to, for example, a shower is likely to become live by contact then it would be an exposed-c-p and should be earthed.


I'm out of my depth geoff, I will have to leave it between you and graham K
 19 April 2018 09:57 PM
User is offline View Users Profile Print this message



AJJewsbury

Posts: 17079
Joined: 13 August 2003

Andy, thanks for the reply, but pipes are allowed as earth electrodes let alone protective conductors, which seems to me that they generally qualify as electical equipment especially where a fault will cause them to become live?

I can see where you're coming from - but I'd suggest not continuing down that road (there be Dragons!). If you carry on like that we'll end up with no main bonding and puny c.p.c.s trying to do the job of PME bonding conductors - with 'undesirable' results.

Like I've said before, the definitions in this area are, to my mind, far from clear and could benefit from a good overhaul - but that thinking is taking us in quite the wrong direction I feel.

- Andy.
 20 April 2018 05:33 PM
User is offline View Users Profile Print this message



gkenyon

Posts: 5159
Joined: 06 May 2002

Originally posted by: Weirdbeard2
Hi GK, if the pipes are earthed by a protective conductor they are exposed conductive parts as they will become live in the event of a fault, they cannot be both extraneous and exposed conductive parts at the same time?
No, the boiler in the example has exposed-conductive-parts to which the pipes connect. The requirements of BS 7671 relating to exposed-conductive-parts applies to the boiler's exposed-conductive-parts, and not the pipes which connect to it.

Being conductors the pipes can carry whatever potential the exposed-conductive-part is at to another part of the premises.

From the bathroom's perspective, they are "extraneous-conductive-parts of the location" because they are not part of the electrical installation, but may import earth potential into the bathroom.

-------------------------
EUR ING Graham Kenyon CEng MIET TechIOSH
G Kenyon Technology Ltd

Web-Site: www.gkenyontech.com
 20 April 2018 06:28 PM
User is offline View Users Profile Print this message



Weirdbeard2

Posts: 496
Joined: 29 November 2017

Thanks for the reply GK, though if a pipe is connected to a CPC it will be largely at the same potential as all the other Bathroom related ciruits cpcs?
 21 April 2018 09:01 PM
User is offline View Users Profile Print this message



AJJewsbury

Posts: 17079
Joined: 13 August 2003

though if a pipe is connected to a CPC it will be largely at the same potential as all the other Bathroom related ciruits cpcs?

No, that's the problem. When you have a L-PE fault, large currents flow (especially on TN systems) and so you can get a large voltage difference between one end of the c.p.c. (MET say) and the other (fault). The UK tradition of using reduced c.s.a. c.p.c.s only makes that even worse.

Say you have a circuit with Zs of 1 Ohm and Ze of say 0.1 Ohms. So R1+R2 totals 0.9 Ohms - let's say a reduced c.p.c. was used and R1=0.3 Ohms and R2=0.6 Ohms.

During a fault 230V/1 Ohm = 230 amps will flow. The c.p.c. (0.6 Ohms) therefore develops a voltage difference of 0.6 Ohms * 230 amps = 138V along its length.

So between exposed-conductive-part where the fault occurred and the MET, there will be 138V - and that will last for as long as disconnection takes (which could be up to 5s even with a modern installation).

So between exposed-conductive-parts of other circuits (whose c.p.c.s are directly connected to the MET) and pipework that's in contact with the exposed-conductive-part where the fault occurred there will also be the same 138V - even if both of those enter the bathroom (without supplementary bonding).

- Andy.
 22 April 2018 08:03 AM
User is offline View Users Profile Print this message



aligarjon

Posts: 3982
Joined: 09 September 2005

Originally posted by: AJJewsbury

If I can try to summarise the ongoing debate so far ...



If we have a metallic part that cannot introduce a potential from outside the installation but can introduce a potential from outside the bathroom (and all the disconnection & RCD requirements of circuits of the location are satisfied) then we have two options:



1. Either ensure that the part is effectively connected to the main bonding (so satisfying the one remaining condition to allow supplementary bonding to be omitted), or

2. Create a supplementary bonding scheme including all circuits of the bathroom as well as that part.



In both cases, it's not necessarily required to add extra G/Y - provided that the existing c.p.c.s etc satisfy that 50V/Ia requirement - between all points in the bathroom in the case of option 2, or between the part and MET in the case of option 1.



So for instance if we had a heating pipe that wasn't directly main bonded, but was in contact with say a boiler c.p.c. - we might consider the worst case fault in this particular instance that could make that pipe live would be a fault on the ring just before the boiler FCU. If the ring was protected by a 32A B-type MCB alone, we'd then be looking at an Ia of 160A, and so R<= 0.3125 Ohms.



If that test failed, we'd either have to run a 10mm2 (say) to the part from the MET or 4mm2 (say) between the part and all c.p.c.s of the bathroom.



- Andy.


Thanks for the summary Andy. Ive just sat here reading the whole thread for a good half an hour and my head is spinning.

The thought that always occurs to me is that it is difficult to omit supplementary bonding in any circumstances in my eyes with the common practice of installing plastic piping in the middle of copper installations, so effectively disconnecting the bathroom pipework from the main bonding. ( This being the same effect of having no bond to a plastic incomer with copper inside that was discussed earlier) . The insertion of plastic piping at some stage in an installations life is predictable these days and we should be allowing for it ?

The other thought that occurred to me was that Central heating pipes going into the ground. There could be a case of bonding at both ends if bonding is required, where it enters and exits the ground. A little like when we have multiple stop taps, clearly on the same incoming pipe somewhere but popping up in different parts of the building. Thats a lot of G/Y .



Gary

-------------------------
Specialised Subject. The Bleedin Obvious. John Cleese
 22 April 2018 10:14 AM
User is offline View Users Profile Print this message



chrispearson

Posts: 434
Joined: 15 February 2018

Originally posted by: AJJewsbury

Say you have a circuit with Zs of 1 Ohm and Ze of say 0.1 Ohms. So R1+R2 totals 0.9 Ohms - let's say a reduced c.p.c. was used and R1=0.3 Ohms and R2=0.6 Ohms.

During a fault 230V/1 Ohm = 230 amps will flow. The c.p.c. (0.6 Ohms) therefore develops a voltage difference of 0.6 Ohms * 230 amps = 138V along its length.

So between exposed-conductive-part where the fault occurred and the MET, there will be 138V - and that will last for as long as disconnection takes (which could be up to 5s even with a modern installation).

So between exposed-conductive-parts of other circuits (whose c.p.c.s are directly connected to the MET) and pipework that's in contact with the exposed-conductive-part where the fault occurred there will also be the same 138V - even if both of those enter the bathroom (without supplementary bonding).


That's the scary bit, which has been making me nervous of taking a shower!

In my bathroom, there is (unsurprisingly) a lighting circuit, which has no exposed CPs, so that's not the risk. There is a boiler whose exposed CPs are probably in continuity with the pipes and hence the towel rail, but I must check. And then there is the fan above the bath/shower. I can just about reach the shower controls with one hand whilst poking a finger of the other hand into the fan, so that's the risk. I must have a look for supplementary bonding there.

(There are no RCDs in the CU.)
 22 April 2018 10:22 AM
User is offline View Users Profile Print this message



Weirdbeard2

Posts: 496
Joined: 29 November 2017

Originally posted by: AJJewsbury

though if a pipe is connected to a CPC it will be largely at the same potential as all the other Bathroom related ciruits cpcs?


No, that's the problem. When you have a L-PE fault, large currents flow (especially on TN systems) and so you can get a large voltage difference between one end of the c.p.c. (MET say) and the other (fault). The UK tradition of using reduced c.s.a. c.p.c.s only makes that even worse.

Say you have a circuit with Zs of 1 Ohm and Ze of say 0.1 Ohms. So R1+R2 totals 0.9 Ohms - let's say a reduced c.p.c. was used and R1=0.3 Ohms and R2=0.6 Ohms.

During a fault 230V/1 Ohm = 230 amps will flow. The c.p.c. (0.6 Ohms) therefore develops a voltage difference of 0.6 Ohms * 230 amps = 138V along its length.

So between exposed-conductive-part where the fault occurred and the MET, there will be 138V - and that will last for as long as disconnection takes (which could be up to 5s even with a modern installation).

So between exposed-conductive-parts of other circuits (whose c.p.c.s are directly connected to the MET) and pipework that's in contact with the exposed-conductive-part where the fault occurred there will also be the same 138V - even if both of those enter the bathroom (without supplementary bonding).



Many thanks for your time Andy, really appreciate it, but if they are both at 138V then the potential between them will be 0V?
 23 April 2018 10:26 AM
User is offline View Users Profile Print this message



AJJewsbury

Posts: 17079
Joined: 13 August 2003

but if they are both at 138V then the potential between them will be 0V?

They would if there were, but they're not. That 138V is between the fault and the MET.

To flesh things out more, let's say Ze = 0.1 Ohm breaks down as 0.05 Ohms each for the supplier's line and earth/PEN, our 230A fault current then gives us 11.5V along the supplier's protective conductor - i.e. between the MET and true earth.

That gives us:

MET: 11.5V above true earth
Faulty appliance (outside the bathroom): 11.5+138V = 149.5V above true earth.

c.p.c.s to other (non-faulty) circuits in the bathroom have no current flowing along them, so remain at the MET's potential (11.5V)
Pipework in contact with the faulty appliance is held at 149.5V

So someone in the bathroom touching an exposed-conductive-part connected to a healthy circuit and the pipework in contact with the faulty circuit feels 138V between them.

- Andy.
 23 April 2018 08:02 PM
User is offline View Users Profile Print this message



Weirdbeard2

Posts: 496
Joined: 29 November 2017

Thanks Andy, though I am lost, you and GK appear to be demonstrating that bathroom supplementary bonding cannot ever be omitted, I was under the impression that It was permissible where all circuit are 30mA RCD protected and are connected together via the installations earthing system, which could just be 2 cpcs such as a light and a shower circuit , I appear to be missing how a bit of pipe also connected to the same CPC system needs treating differently to the cpc that happens to be part of a cable?
IET » Wiring and the regulations » Determining an extraneous conductive-part by measurement

<< 1 2 3 4 5 6 Previous Next Last unread
Topic Tools Topic Tools
Statistics

New here?


See Also:



FuseTalk Standard Edition v3.2 - © 1999-2018 FuseTalk Inc. All rights reserved.

 
..