IET logo
 
IET
Decrease font size
Increase font size
Topic Title: EARTH FAULT LOOP IMPEDANCE TESTING
Topic Summary: MAIN PROTECTIVE BONDING CONDUCTORS
Created On: 31 December 2012 11:42 AM
Status: Post and Reply
Linear : Threading : Single : Branch
1 2 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.
 31 December 2012 11:42 AM
User is offline View Users Profile Print this message



John Peckham

Posts: 7360
Joined: 23 April 2005

Given that it is widely accepted, apart from one 2nd rate text book, that it is hazardous/dangerous to users of the installation to carry out EFLI tests with bonding conductors disconnected surely on the same basis it must be hazardous to carry out EFLI tests on installations that do not have main bonds?

So having carried out your visual inspection and discovered, a very common situation, that there are no bonds to any or all extraneous conductive parts should you then proceed to carry out EFLI testing?

Assuming, as you should be, using a modern EFLI tester to BS EN 61557-3 the test current will automatically cut of if the protective conductor rises to 50V and a time up to 40ms can we rely on this safety feature to test when no bonding is present?

If we go ahead and test without bonding how valid are the test results assuming bonding will be fitted after you have submitted your "unsatisfactory" EICR to the client (some hope)?

I am not advocating the unsafe practice of removing bonding conductors prior to EFLI testing!

-------------------------
John Peckham

http://www.astutetechnicalservices.co.uk/
 31 December 2012 12:52 PM
User is offline View Users Profile Print this message


Avatar for spinlondon.
spinlondon

Posts: 4437
Joined: 10 December 2004

There are a number of activities which are or could be hazardous.
Hazards can be removed or at the very least reduced if a safe method of operation is adopted.
I would not consider conducting an EFLI test to be particularly hazardous, if the premises which contains the installation is empty.

Then of course, the Ze test is an EFLI test.
 31 December 2012 02:40 PM
User is offline View Users Profile Print this message


Avatar for Parsley.
Parsley

Posts: 960
Joined: 04 November 2004

John

I've tried to work it out using values from GN 5 figure 13.2 for a TNC-S system

If = 230V/ the testers ELFI resistance (20 ohms I believe for the fluke 1651) + ze 0.35 + R1 0.5 +R2 0.5 = 230/21.35= 10.77 A


Ut With bonding = If x R2
10.77 x 0.5 = 5.38V

Ut Without bonding = voltage drop in PEN = 10.77 x 0.175= 1.88V + 5.38V =7.26V

I think If the PEN was actually loaded the Ut would be higher.

Ut would be a lot higher in a TT system, I get 0.81V with bonding and 163.81V without.

This might all be wrong, but I'm stuck in the office today with not much going on so thought I would have a go.

Happy New year
 31 December 2012 03:44 PM
User is offline View Users Profile Print this message



AJJewsbury

Posts: 11218
Joined: 13 August 2003

I agree with Parsley, in terms of shock protection from faults (or tests) internal to the installation, bonding makes only a very small difference on properly functioning TN systems - the low Ze means that most of the voltage is divided across the circuit conductors - especially reduced csa c.p.c. - and bonding doesn't mitigate that much. Different for TT of course - and for PME bonding is essential insurance against a broken supply CNE.

If we're testing we've got to consider the possibility that what we're testing could be faulty. You could have all bonds in place & proved but then test a final circuit that had a broken c.p.c. - the test might then put a lethal voltage on that part of the installation downstream of the break.

I guess if you want to be really safe, you need to prove all necessary continuity first - with dead tests - e.g. wander lead R1 test - (R1+R2 isn't any good if you have to mess around with the connections after testing as there's always the possibility of faulty reconnection). Then just use the loop tester to "quantify" the result.

As ever, it's down to risk-cost balance - is it worth dead testing every point (e.g. on a periodic) first - in case of a 1 in a million chance that some harm might occur from the loop test.

- Andy.
 01 January 2013 10:12 AM
User is offline View Users Profile Print this message



lyledunn

Posts: 603
Joined: 13 August 2003

Going way back, GN3 advocated EFLI tests with main bonding disconnected. Presumably this was to discount the effect of the bonds on the test results. It was also a question in the Level 3 C and G 2391 and installation papers. I wrote to both parties to indicate the folly of having main protective bonding disconnected whilst the installation remained live. Not so much for any hazard caused by the test instrument but more from the fact that it was not beyond the bounds of possibility that an earth fault could occur during the period that the bonding was disconnected. I am sure that I was not t he only one to have pointed that out, but in any event both parties changed tact.

-------------------------
Regards,

Lyle Dunn
 01 January 2013 10:34 AM
User is offline View Users Profile Print this message



John Peckham

Posts: 7360
Joined: 23 April 2005

Lyle

You are correct my 1997 edition of GN3 does say that bonding conductors should be disconnected for EFLI testing. Clearly the great and the good of the IEE/IET have changed their guidance on this practice.

-------------------------
John Peckham

http://www.astutetechnicalservices.co.uk/
 01 January 2013 10:45 AM
User is online View Users Profile Print this message


Avatar for GeoffBlackwell.
GeoffBlackwell

Posts: 3487
Joined: 18 January 2003

The 1997 error in 3.9.2 was corrected by an Errata sheet published 15/02/2000.

Note that I left the NICEIC in 1995 and things started going to pot around that time .

Regards

Geoff Blackwell
 01 January 2013 05:48 PM
User is offline View Users Profile Print this message



Jaymack

Posts: 4565
Joined: 07 April 2004

The procedures for inspection and testing, call for verification of earthing and bonding as a pre-requisite during an inspection report etc. Although not sacrosant, having found such a departure during verification, is a subsequent test for Zs acceptable?

Further, the notorious visual condition report conceived by the NICEIC, is supposedly just that ...... a visual report, but a further step and not involving much time, could be a test for Zs; at least for socket outlets where the greatest danger exists for Joe Soap. Should that be carried out, in light of the requirement for the verification as above?

Regards
 02 January 2013 09:37 AM
User is offline View Users Profile Print this message


Avatar for Legh.
Legh

Posts: 3472
Joined: 17 December 2004

Not to be too contentious, just a few questions....

1/ What are the probabilities of a dangerous occurrence happening when applying a loop test at 15mA for 40ms on a TT system with a Ze of, lets say, 200Ω, when all bonding is disconnected ?

2/ Is this rule aimed at installations that could be still live in parts or smaller installations where the inspector has a greater degree of control over access ?
and
secondly why would you need to do this test in the first case?

Surely a Ze test and a R1 + R2 low resistance test would be sufficient if there was the remotest possibility of a non-continuous cpc?

3/ How can a recommendation in 1997 GN3 for bonding to be disconnected when conducting an earth loop test then suddenly become an error in 2000?

I think this is highly theoretical and risks are assessed in a board room with too much tea and biscuits.

So does anybody have any real situations and are there any 'real statistics' to prove that this is dangerous ?

Legh

-------------------------
Why do we need Vernier Calipers when we have container ships?

http://www.leghrichardson.co.uk

"Science has overcome time and space. Well, Harvey has overcome not only time and space - but any objections."
 02 January 2013 09:56 AM
User is online View Users Profile Print this message


Avatar for GeoffBlackwell.
GeoffBlackwell

Posts: 3487
Joined: 18 January 2003

3/ How can a recommendation in 1997 GN3 for bonding to be disconnected when conducting an earth loop test then suddenly become an error in 2000?


GN3 1992 & 1995 did not include the proposal that bonding was removed. The 1997 version did and many objections were raised. This resulted in the Errata sheet published 15/02/2000.

On the question of danger from any test - this must be assessed by the inspector for the given situation. I have insisted that around 30 very expensive horses be removed from a stable before I would test the RCDs - they just took them out for run .

Regards

Geoff Blackwell
 02 January 2013 10:13 AM
User is offline View Users Profile Print this message


Avatar for Parsley.
Parsley

Posts: 960
Joined: 04 November 2004

I think JP might be referring to EICR's where live ELFI testing is used to confirm earth continuity, Often used for convenience and cost implications.

As per note 4 of table 3.4 GN3

The earth fault loop impedance test may be used to confirm the continuity of protective conductors at socket-outlets and at accessible exposed-conductive-parts of current-using equipment and accessories." .

Regards
 02 January 2013 11:07 AM
User is offline View Users Profile Print this message


Avatar for Legh.
Legh

Posts: 3472
Joined: 17 December 2004

Originally posted by: Parsley

I think JP might be referring to EICR's where live ELFI testing is used to confirm earth continuity, Often used for convenience and cost implications.

As per note 4 of table 3.4 GN3

The earth fault loop impedance test may be used to confirm the continuity of protective conductors at socket-outlets and at accessible exposed-conductive-parts of current-using equipment and accessories." .


Interesting.....

Now some chap with a mug of tea in one hand and a slice of lardy cake in the other decided, after weighing up all the probabilities, it would be perfectly acceptable to 'confirm the continuity of the protective conductors at various points where there are exposed conductive parts' by using an earth fault loop impedance tester.....

So in fact there is relatively little danger since confirmation is no guaranty of a definite continuity of R2 ? ...


Legh

-------------------------
Why do we need Vernier Calipers when we have container ships?

http://www.leghrichardson.co.uk

"Science has overcome time and space. Well, Harvey has overcome not only time and space - but any objections."
 02 January 2013 11:25 AM
User is offline View Users Profile Print this message



John Peckham

Posts: 7360
Joined: 23 April 2005

Legh

"So in fact there is relatively little danger since confirmation is no guaranty of a definite continuity of R2 ? ... "

I think we have debated many times on this forum there is no test that will prove continuity of R2 beyond doubt. The only method would be visual inspection throughout the entire run of the conductor which is not practical for anything other than the simplest installation. Then there is the question of what use an R2 test is given the test is at 200mA but the fault current may be many hundreds of amps or thousands of amps. GBs golden spanner short circuit test will test the integrity of the earth fault path at least for the duration of the test!

A loop test at high current (25A) is a more robust test of earth continuity but a poor substitute for the golden spanner.

-------------------------
John Peckham

http://www.astutetechnicalservices.co.uk/
 02 January 2013 11:38 AM
User is offline View Users Profile Print this message


Avatar for Parsley.
Parsley

Posts: 960
Joined: 04 November 2004

When carrying out an EICR It's very difficult to prove the R2 of a circuit in a commercial installation especially with metallic containment systems and the associated parallel paths. If you try to prove it using a wander lead it won't be accurate either unless all the parallel paths are removed, which is likely to be impossible and far to time consuming in any case on most commercial/industrial buildings and introduces the risk of damaging the exisitng terminations equipment etc?

3.8.1 of GN3 states.


"The purpose of periodic inspection and testing is to provide an engineering view on whether or not the installation is in a satisfactory condition where it can continue to be used in a safe way.
The periodic inspection and test comprises a detailed examination of the installation together with appropriate tests. The inspection is carried out without taking apart or dismantling equipment as far as is possible. The tests made are mainly to confirm that the disconnection times stated in Chapter 41 are met, as well as highlighting other defects."


Regards
 02 January 2013 12:26 PM
User is offline View Users Profile Print this message



AJJewsbury

Posts: 11218
Joined: 13 August 2003

When carrying out an EICR It's very difficult to prove the R2 of a circuit in a commercial installation especially with metallic containment systems and the associated parallel paths. If you try to prove it using a wander lead it won't be accurate either unless all the parallel paths are removed, which is likely to be impossible and far to time consuming in any case on most commercial/industrial buildings and introduces the risk of damaging the exisitng terminations equipment etc?

Although in terms of checking the safety of the installation at the time of the test is it so bad to make use of parallel paths? - the fault current ain't fussy after all - it'll go through a steel stanchion as happily as SWA. OK there's doubt about what might happen in the future should there be changes in the building, but to some extent we have to live with that anyway (what happens if someone mends a rewireable fuse with the wrong size wire or the DNO dig a hole in the street and changes Ze) - and it might be 'less obvious' that "non-electrical" changes might impact on the safety of the electrical system, but fundamentally, like an MOT, the test results are only true at the moment of the test and guarantee nothing for the future.

If an R2 test shows continuity, then at the time of the test, earth continuity is good. It's not ideal, but provided everyone is aware of the limitations of the real world, we should be able to live with it.

- Andy.
 02 January 2013 12:27 PM
User is offline View Users Profile Print this message


Avatar for Legh.
Legh

Posts: 3472
Joined: 17 December 2004

Originally posted by: John Peckham

Legh

"So in fact there is relatively little danger since confirmation is no guaranty of a definite continuity of R2 ? ... "

I think we have debated many times on this forum there is no test that will prove continuity of R2 beyond doubt. The only method would be visual inspection throughout the entire run of the conductor which is not practical for anything other than the simplest installation.
Agreed, there is no guarantee that even after a thorougher visual inspection of any cable for any installation that there is likely to be 100% confidence of continuity of the R2

Then there is the question of what use an R2 test is given the test is at 200mA but the fault current may be many hundreds of amps or thousands of amps. GBs golden spanner short circuit test will test the integrity of the earth fault path at least for the duration of the test!

200mA doesn't prove reliability and possible energy let through. This brings us back to proving the integrity of R2 as opposed to possible reliance on parallel paths which may suggest that for individual circuit checking TH may have had a point. Its still theoretical and unrealistic to do this for every circuit and not an option where the installation is in continual use. You could only offer a 95% confidence level at best.

Under real faults with large currents you would be in the hands of the protective device which would suggest that the maximum Zs was not compromised and the breaking capacity of an upstream or the cct breaker was acceptable for the minimum Zs


A loop test at high current (25A) is a more robust test of earth continuity but a poor substitute for the golden spanner.


It sounds like a better method than no testing at all although I would have thought that Geoff's collection of 'Golden Spanners' would be better placed in investment rather than gold plating the insides of switchgear cabinets each time 'a goodness of connection/continuity' was required

@parsley
I couldn't agree more the less you remove the better which in effect means you are reliant on the established earthing regardless of the path the fault current is likely to take. Its not perfect but then what is and becomes the experience and view of the inspector.

Just to move sideways a little.
To me, initial testing is definitely the place to start. (which leads onto competent initial design, installation and testing, not to open up that old can of worms and all its variables), R2 results on EICs could then be verified where all the design parameters are placed infront of the inspector rather than continually re-inventing the wheel with 'Golden Spanners'

Legh

-------------------------
Why do we need Vernier Calipers when we have container ships?

http://www.leghrichardson.co.uk

"Science has overcome time and space. Well, Harvey has overcome not only time and space - but any objections."
 02 January 2013 12:44 PM
User is offline View Users Profile Print this message


Avatar for Parsley.
Parsley

Posts: 960
Joined: 04 November 2004

Originally posted by: Legh

Originally posted by: John Peckham



Legh



"So in fact there is relatively little danger since confirmation is no guaranty of a definite continuity of R2 ? ... "



I think we have debated many times on this forum there is no test that will prove continuity of R2 beyond doubt. The only method would be visual inspection throughout the entire run of the conductor which is not practical for anything other than the simplest installation.

Agreed, there is no guarantee that even after a thorougher visual inspection of any cable for any installation that there is likely to be 100% confidence of continuity of the R2



Then there is the question of what use an R2 test is given the test is at 200mA but the fault current may be many hundreds of amps or thousands of amps. GBs golden spanner short circuit test will test the integrity of the earth fault path at least for the duration of the test!



200mA doesn't prove reliability and possible energy let through. This brings us back to proving the integrity of R2 as opposed to possible reliance on parallel paths which may suggest that for individual circuit checking TH may have had a point. Its still theoretical and unrealistic to do this for every circuit and not an option where the installation is in continual use. You could only offer a 95% confidence level at best.



Under real faults with large currents you would be in the hands of the protective device which would suggest that the maximum Zs was not compromised and the breaking capacity of an upstream or the cct breaker was acceptable for the minimum Zs




A loop test at high current (25A) is a more robust test of earth continuity but a poor substitute for the golden spanner.




It sounds like a better method than no testing at all although I would have thought that Geoff's collection of 'Golden Spanners' would be better placed in investment rather than gold plating the insides of switchgear cabinets each time 'a goodness of connection/continuity' was required



@parsley

I couldn't agree more the less you remove the better which in effect means you are reliant on the established earthing regardless of the path the fault current is likely to take. Its not perfect but then what is and becomes the experience and view of the inspector.



Just to move sideways a little.

To me, initial testing is definitely the place to start. (which leads onto competent initial design, installation and testing, not to open up that old can of worms and all its variables), R2 results on EICs could then be verified where all the design parameters are placed infront of the inspector rather than continually re-inventing the wheel with 'Golden Spanners'



Legh


I don't generally see many EIC's with R2 values entered, it's normally the R1+R2 box that gets completed, probably because that method has also been used to verify polarity in theory.

Regards
 02 January 2013 12:52 PM
User is offline View Users Profile Print this message


Avatar for Legh.
Legh

Posts: 3472
Joined: 17 December 2004

I don't generally see many EIC's with R2 values entered, it's normally the R1+R2 box that gets completed, probably because that method has also been used to verify polarity in theory.


Lol, I suspect I was talking figuratively in that in measuring 'R2' presupposes any method which might give a reasonable R2 value. Albeit, ambiguous I also suspect ..... Its starting raining again ....

Legh

-------------------------
Why do we need Vernier Calipers when we have container ships?

http://www.leghrichardson.co.uk

"Science has overcome time and space. Well, Harvey has overcome not only time and space - but any objections."
 02 January 2013 02:36 PM
User is offline View Users Profile Print this message



AJJewsbury

Posts: 11218
Joined: 13 August 2003

it's normally the R1+R2 box that gets completed, probably because that method has also been used to verify polarity in theory.

If only it did. It can't spot R1/R2 reversal - so only verifies polarity if R2 is already known to be correctly connected and only proves R2 continuity if R1 is already known to be correct. Chicken & egg.
- Andy.
 02 January 2013 02:55 PM
User is offline View Users Profile Print this message


Avatar for ebee.
ebee

Posts: 5661
Joined: 02 December 2004

I think we might consult the expert David Cockburn on this subject

-------------------------
Regards,
Ebee (M I S P N)

Knotted cables cause Lumpy Lektrik
IET » Wiring and the regulations » EARTH FAULT LOOP IMPEDANCE TESTING

1 2 Next Last unread
Topic Tools Topic Tools
Statistics

See Also:



FuseTalk Standard Edition v3.2 - © 1999-2014 FuseTalk Inc. All rights reserved.