IET logo
 
IET
Decrease font size
Increase font size
Topic Title: Insulation resistance
Topic Summary: If we accept that R1 R2 testing has issues?
Created On: 19 May 2011 10:20 PM
Status: Post and Reply
Linear : Threading : Single : Branch
 Insulation resistance   - kj scott - 19 May 2011 10:20 PM  
 Insulation resistance   - JonSteward - 20 May 2011 06:55 AM  
 Insulation resistance   - sparkingchip - 20 May 2011 08:45 AM  
 Insulation resistance   - AJJewsbury - 20 May 2011 08:51 AM  
 Insulation resistance   - AJJewsbury - 20 May 2011 03:48 PM  
 Insulation resistance   - JonSteward - 20 May 2011 04:12 PM  
 Insulation resistance   - Paradigm - 20 May 2011 04:50 PM  
 Insulation resistance   - JonSteward - 20 May 2011 05:13 PM  
 Insulation resistance   - kj scott - 20 May 2011 07:16 PM  
 Insulation resistance   - slittle - 20 May 2011 07:59 PM  
 Insulation resistance   - John Peckham - 20 May 2011 08:30 PM  
 Insulation resistance   - kj scott - 20 May 2011 08:57 PM  
 Insulation resistance   - GeoffBlackwell - 20 May 2011 09:02 PM  
 Insulation resistance   - rocknroll - 20 May 2011 09:05 PM  
 Insulation resistance   - kj scott - 20 May 2011 09:10 PM  
 Insulation resistance   - timothyboler - 22 May 2011 11:20 AM  
 Insulation resistance   - MBE - 22 May 2011 04:17 PM  
 Insulation resistance   - sparkingchip - 22 May 2011 04:26 PM  
 Insulation resistance   - Zs - 23 May 2011 08:27 PM  
 Insulation resistance   - kj scott - 25 May 2011 11:13 AM  
 Insulation resistance   - timothyboler - 25 May 2011 12:20 PM  
 Insulation resistance   - kj scott - 25 May 2011 01:24 PM  
 Insulation resistance   - rocknroll - 25 May 2011 01:46 PM  
 Insulation resistance   - OMS - 25 May 2011 02:00 PM  
 Insulation resistance   - kj scott - 25 May 2011 03:09 PM  
 Insulation resistance   - rocknroll - 25 May 2011 03:29 PM  
 Insulation resistance   - kj scott - 25 May 2011 06:12 PM  
 Insulation resistance   - OMS - 25 May 2011 06:44 PM  
 Insulation resistance   - kj scott - 25 May 2011 06:52 PM  
 Insulation resistance   - OMS - 25 May 2011 06:59 PM  
 Insulation resistance   - kj scott - 25 May 2011 07:21 PM  
 Insulation resistance   - rocknroll - 25 May 2011 07:00 PM  
 Insulation resistance   - OMS - 25 May 2011 07:11 PM  
 Insulation resistance   - rocknroll - 25 May 2011 07:23 PM  
 Insulation resistance   - Jaymack - 25 May 2011 07:05 PM  
 Insulation resistance   - OMS - 25 May 2011 07:28 PM  
 Insulation resistance   - OMS - 25 May 2011 07:30 PM  
 Insulation resistance   - kj scott - 25 May 2011 07:44 PM  
 Insulation resistance   - GeoffBlackwell - 25 May 2011 08:22 PM  
 Insulation resistance   - kj scott - 25 May 2011 08:26 PM  
 Insulation resistance   - Jaymack - 25 May 2011 08:59 PM  
 Insulation resistance   - kj scott - 25 May 2011 09:01 PM  
 Insulation resistance   - kj scott - 25 May 2011 09:17 PM  
 Insulation resistance   - Jaymack - 25 May 2011 09:32 PM  
 Insulation resistance   - kj scott - 26 May 2011 10:08 AM  
 Insulation resistance   - dg66 - 26 May 2011 10:40 AM  
 Insulation resistance   - timothyboler - 26 May 2011 12:26 PM  
 Insulation resistance   - GeoffBlackwell - 26 May 2011 01:23 PM  
 Insulation resistance   - timothyboler - 26 May 2011 01:39 PM  
 Insulation resistance   - GeoffBlackwell - 26 May 2011 01:59 PM  
 Insulation resistance   - kj scott - 26 May 2011 02:43 PM  
 Insulation resistance   - mikejumper - 20 May 2011 09:21 PM  
 Insulation resistance   - kj scott - 20 May 2011 09:30 PM  
 Insulation resistance   - sparkingchip - 20 May 2011 09:31 PM  
 Insulation resistance   - slittle - 20 May 2011 09:38 PM  
 Insulation resistance   - kj scott - 20 May 2011 09:52 PM  
 Insulation resistance   - slittle - 20 May 2011 09:55 PM  
 Insulation resistance   - sparkingchip - 20 May 2011 09:44 PM  
 Insulation resistance   - JonSteward - 20 May 2011 09:47 PM  
 Insulation resistance   - sparkingchip - 20 May 2011 09:54 PM  
 Insulation resistance   - kj scott - 20 May 2011 09:58 PM  
 Insulation resistance   - kj scott - 20 May 2011 10:00 PM  
 Insulation resistance   - sparkingchip - 20 May 2011 10:03 PM  
 Insulation resistance   - kj scott - 20 May 2011 10:06 PM  
 Insulation resistance   - slittle - 20 May 2011 10:06 PM  
 Insulation resistance   - sparkingchip - 20 May 2011 10:22 PM  
 Insulation resistance   - kj scott - 20 May 2011 10:26 PM  
 Insulation resistance   - prtadko - 20 May 2011 10:44 PM  
 Insulation resistance   - kj scott - 20 May 2011 10:55 PM  
 Insulation resistance   - Maxpower - 20 May 2011 10:54 PM  
 Insulation resistance   - sparkingchip - 21 May 2011 12:03 PM  
 Insulation resistance   - sparkingchip - 25 May 2011 11:38 PM  
 Insulation resistance   - Jaymack - 26 May 2011 08:57 AM  
 Insulation resistance   - AJJewsbury - 26 May 2011 01:35 PM  
 Insulation resistance   - AJJewsbury - 26 May 2011 02:21 PM  
 Insulation resistance   - GeoffBlackwell - 26 May 2011 02:30 PM  
 Insulation resistance   - kj scott - 26 May 2011 02:47 PM  
 Insulation resistance   - davezawadi - 26 May 2011 02:57 PM  
 Insulation resistance   - GeoffBlackwell - 26 May 2011 03:13 PM  
 Insulation resistance   - AJJewsbury - 26 May 2011 02:38 PM  
 Insulation resistance   - AJJewsbury - 26 May 2011 03:27 PM  
 Insulation resistance   - kj scott - 26 May 2011 04:05 PM  
 Insulation resistance   - AJJewsbury - 26 May 2011 04:20 PM  
 Insulation resistance   - kj scott - 26 May 2011 04:27 PM  
 Insulation resistance   - AJJewsbury - 26 May 2011 04:34 PM  
 Insulation resistance   - kj scott - 26 May 2011 04:56 PM  
 Insulation resistance   - Jaymack - 26 May 2011 07:05 PM  
 Insulation resistance   - OMS - 26 May 2011 07:17 PM  
 Insulation resistance   - kj scott - 26 May 2011 07:38 PM  
 Insulation resistance   - Jaymack - 26 May 2011 08:08 PM  
 Insulation resistance   - ebee - 27 May 2011 07:50 AM  
 Insulation resistance   - kj scott - 27 May 2011 01:43 PM  
 Insulation resistance   - kj scott - 27 May 2011 01:33 PM  
 Insulation resistance   - Jaymack - 27 May 2011 02:11 PM  
 Insulation resistance   - kj scott - 27 May 2011 02:42 PM  
 Insulation resistance   - Jaymack - 27 May 2011 03:10 PM  
 Insulation resistance   - kj scott - 27 May 2011 03:40 PM  
 Insulation resistance   - Jaymack - 27 May 2011 04:19 PM  
 Insulation resistance   - kj scott - 27 May 2011 04:50 PM  
 Insulation resistance   - Jaymack - 27 May 2011 05:42 PM  
 Insulation resistance   - kj scott - 27 May 2011 09:54 PM  
 Insulation resistance   - ebee - 27 May 2011 10:22 PM  
 Insulation resistance   - Jobbo - 27 May 2011 10:54 PM  
 Insulation resistance   - kj scott - 27 May 2011 11:37 PM  
 Insulation resistance   - GeoffBlackwell - 28 May 2011 06:17 AM  
 Insulation resistance   - ebee - 28 May 2011 06:47 AM  
 Insulation resistance   - OMS - 28 May 2011 09:48 AM  
 Insulation resistance   - Jaymack - 28 May 2011 09:50 AM  
 Insulation resistance   - ebee - 28 May 2011 10:02 AM  
 Insulation resistance   - kj scott - 27 May 2011 11:36 PM  
 Insulation resistance   - Jaymack - 28 May 2011 09:39 AM  
 Insulation resistance   - GeoffBlackwell - 28 May 2011 09:42 AM  
 Insulation resistance   - kj scott - 28 May 2011 11:26 AM  
 Insulation resistance   - Jaymack - 28 May 2011 12:21 PM  
 Insulation resistance   - sparkingchip - 28 May 2011 11:43 AM  
 Insulation resistance   - kj scott - 28 May 2011 11:59 AM  
 Insulation resistance   - kj scott - 28 May 2011 12:24 PM  
 Insulation resistance   - ant1uk - 28 May 2011 12:00 PM  
 Insulation resistance   - OMS - 28 May 2011 12:14 PM  
 Insulation resistance   - ant1uk - 28 May 2011 12:19 PM  
 Insulation resistance   - OMS - 28 May 2011 12:23 PM  
 Insulation resistance   - ant1uk - 28 May 2011 12:29 PM  
 Insulation resistance   - davezawadi - 29 May 2011 08:05 AM  
 Insulation resistance   - Phoenix151 - 29 May 2011 08:41 AM  
 Insulation resistance   - John Peckham - 29 May 2011 10:15 AM  
 Insulation resistance   - dg66 - 29 May 2011 11:43 AM  
 Insulation resistance   - Phoenix151 - 29 May 2011 12:03 PM  
 Insulation resistance   - dg66 - 29 May 2011 12:22 PM  
 Insulation resistance   - sparkingchip - 29 May 2011 12:24 PM  
 Insulation resistance   - John Peckham - 29 May 2011 06:03 PM  
 Insulation resistance   - Zs - 29 May 2011 07:01 PM  
 Insulation resistance   - kj scott - 07 June 2011 08:20 PM  
 Insulation resistance   - AJJewsbury - 31 May 2011 12:49 PM  
 Insulation resistance   - lyledunn - 02 June 2011 05:23 PM  
 Insulation resistance   - OMS - 03 June 2011 01:13 PM  
 Insulation resistance   - lyledunn - 07 June 2011 11:51 PM  
 Insulation resistance   - kj scott - 13 June 2011 09:58 AM  
 Insulation resistance   - OMS - 13 June 2011 10:17 AM  
 Insulation resistance   - kj scott - 07 June 2011 08:26 PM  
 Insulation resistance   - sparkingchip - 04 June 2011 09:18 PM  
Search Topic Search Topic
Topic Tools Topic Tools
View similar topics View similar topics
View topic in raw text format. Print this topic.
 19 May 2011 10:20 PM
User is offline View Users Profile Print this message


Avatar for kj scott.
kj scott

Posts: 2144
Joined: 02 April 2006

Accepting, that the validity is questionable of R1,R2 testing, and therefore ring continuity: how about.
Is the insulation resistance test valid, or even viable?
What does it prove?
Can you realistically apply the test to a whole installation; as prescribed by 612.3.2; if you did would the test result be valid?

-------------------------
http://www.niceic.biz
 20 May 2011 06:55 AM
User is offline View Users Profile Print this message


Avatar for JonSteward.
JonSteward

Posts: 567
Joined: 04 December 2007

As debated before.
Clamp meters and thermal imaging as an alternative to IR testing in large installations has been mooted. IR testing does have its place on new circuit installs and an aid to fault finding.
 20 May 2011 08:45 AM
User is offline View Users Profile Print this message


Avatar for sparkingchip.
sparkingchip

Posts: 5973
Joined: 18 January 2003

If all the sockets are working, despite failing ring continuity, then maybe.

If there is not continuity on the L+N but the sockets are working then you can assume one break on one section of cable, so insulation test should be valid. But if there is a break on the CPC, the sockets could work with multiple faults thus invalidating the insulation test as there could be two breaks on the CPC and the disconnected section could be subject to mains voltage due to a insulation failure.

Testing individually at sockets with the loop tester after continuity test is a better bet to ensure the insulation tester is reaching all parts of the circuit, there could be a twisted pair fell out of the back of a faulty socket.

I once tested a brand new install by someone else, there was a neon on the cooker switch, so I dropped it out and tested the two halves of the circuit, reconnected the brand new never used fitting (major brand) and turned the power on to see the switch blow up, you never can tell !

The cable in my present Avatar can pass a insulation test.

Teaching grandma to suck eggs is the expression that comes to mind here!

Andy
 20 May 2011 08:51 AM
User is online View Users Profile Print this message



AJJewsbury

Posts: 11374
Joined: 13 August 2003

the validity is questionable of R1,R2 testing, and therefore ring continuity

I thought the problem with R1+R2 and R2 testing was due to parallel paths to the c.p.c. - that shouldn't be a problem with live conductors, so the important bit of ring continuity testing - i.e. r1 & rn (to prevent overloading of the intact leg) should be viable. In a way breaks in r2 are less important provided Zs is still adequate (high protective conductor currents notwithstanding).
- Andy.
 20 May 2011 03:48 PM
User is online View Users Profile Print this message



AJJewsbury

Posts: 11374
Joined: 13 August 2003

Clamp meters and thermal imaging as an alternative to IR testing in large installations has been mooted.

I think thermal imaging only shows conductors getting hot - so useful for spotting overloads and bad connections, but I'm not sure it would spot poor insulation (unless it was that bad it was letting huge amount of current leak).

- Andy.
 20 May 2011 04:12 PM
User is offline View Users Profile Print this message


Avatar for JonSteward.
JonSteward

Posts: 567
Joined: 04 December 2007

Originally posted by: AJJewsbury

Clamp meters and thermal imaging as an alternative to IR testing in large installations has been mooted.


I think thermal imaging only shows conductors getting hot - so useful for spotting overloads and bad connections, but I'm not sure it would spot poor insulation (unless it was that bad it was letting huge amount of current leak).



- Andy.


I was thinking along the lines of the 'Gold Standard' PIR'ing where those meters would be essential for the professional PIR tool box. (Along with a telescopic pole, R2 lead, battery and bell, as has been suggested before)
I use an MFT 1502 but don't hold that against me.
 20 May 2011 04:50 PM
User is offline View Users Profile Print this message


Avatar for Paradigm.
Paradigm

Posts: 700
Joined: 10 September 2010

Off topic

Jon Steward. Love your avatar, that is one very weird film, still haven't got over it yet and I saw it over 25 years ago.

Sorry for thread hijack

Nick

-------------------------
"be careful of what you write"
 20 May 2011 05:13 PM
User is offline View Users Profile Print this message


Avatar for JonSteward.
JonSteward

Posts: 567
Joined: 04 December 2007

David Lynch at his very best.
I've just reordered a copy from Amazon to show my kids
 20 May 2011 07:16 PM
User is offline View Users Profile Print this message


Avatar for kj scott.
kj scott

Posts: 2144
Joined: 02 April 2006

Table 61 minimum insulation resistance 1MΩ for whole DB.
What insulation resistance value would you expect to see, L-E and L-N on a single circuit wired in T&E buried in a plastered brick wall, when the cable has a nail driven directly into the line conductor only?

-------------------------
http://www.niceic.biz
 20 May 2011 07:59 PM
User is offline View Users Profile Print this message


Avatar for slittle.
slittle

Posts: 3487
Joined: 22 November 2007

Had one last week, where the kitchen fitter had successfully put a bracket up over two t&e's feeding a socket on the kitchen ring.

One screw (rawlplug) got the N of one t&e, the other got the live of the other one.

The only thing stopping it going bang was the plastic rawlplugs, IR ok.


Stu
 20 May 2011 08:30 PM
User is offline View Users Profile Print this message



John Peckham

Posts: 7418
Joined: 23 April 2005

For us Gold Standard members it's cream cakes all round if we find a large DB that goes over 1M on a commercial PIR L&N to E. Strange how our competitors get >200 for every board and are able to measure the same L to L and L to N. How do they do that?

I saw a previous PIR on a very large central London office block I was pricing last week. For every board the Zs was 0.01ohms except for one kitchen which was 0.08 ohms and every board was >200M . Only 22 faults reported for the whole building and most of those were 3s and 4s. Mind you loads of paper in lovely folders.

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

http://www.astutetechnicalservices.co.uk/
 20 May 2011 08:57 PM
User is offline View Users Profile Print this message


Avatar for kj scott.
kj scott

Posts: 2144
Joined: 02 April 2006

Factory unit last month all >299MΩ for previous report, N-E was 7.9MΩ with socket outlet circuits excluded, due to RCD outlets.
Another unit by same contractor, all previous >299MΩ, I did not do insulation tests, as I could not disconnect N-E paths without major dismantling of the factory, approx two days work, and they only close Sundays, also some isolators were obstructed by machines, so doors don't open.

-------------------------
http://www.niceic.biz
 20 May 2011 09:02 PM
User is offline View Users Profile Print this message


Avatar for GeoffBlackwell.
GeoffBlackwell

Posts: 3516
Joined: 18 January 2003

Well you so called 'Gold Standard Members' had better sharpen up your act and stop using incorrect terminology.

Please read the title of BS 7671 Chapter 62 - you will see that the process is periodic inspection and testing (PIT). The report (PIR) is only part of this process.

Now if you do a number of these the term would be PITS - and some that I have reviewed have definitely been the PITS .

Regards

Geoff Blackwell
 20 May 2011 09:05 PM
User is offline View Users Profile Print this message


Avatar for rocknroll.
rocknroll

Posts: 8838
Joined: 03 October 2005

Well you so called 'Gold Standard Members' had better sharpen up your act and stop using incorrect terminology.


Thats a classic for a Fri night!
LOL

regards

-------------------------
"Take nothing but a picture,
leave nothing but footprints!"
-------------------------
"Oh! The drama of it all."
-------------------------
"You can throw all the philosophy you like at the problem, but at the end of the day it's just basic electrical theory!"
-------------------------
 20 May 2011 09:10 PM
User is offline View Users Profile Print this message


Avatar for kj scott.
kj scott

Posts: 2144
Joined: 02 April 2006

I didn't start this as a whinge about PIT standards, more to consider that the requirement of 612.3.2 is unrealistic, and the minimum value 1MΩ, too low.
The regulation is also contradicted by the note below table 61 which permits 1MΩ for a single circuit.
GN3 has only basic guidance of insulation testing, and a process that will only work on a 2391 test rig.

-------------------------
http://www.niceic.biz
 22 May 2011 11:20 AM
User is offline View Users Profile Print this message


Avatar for timothyboler                                      .
timothyboler

Posts: 230
Joined: 25 July 2008

Originally posted by: kj scott

I didn't start this as a whinge about PIT standards, more to consider that the requirement of 612.3.2 is unrealistic, and the minimum value 1MΩ, too low.

The regulation is also contradicted by the note below table 61 which permits 1MΩ for a single circuit.



Hi, where is the contradiction? I agree that 1MΩ is in the majority of installations too low (GN3 says be worried if it is lower than 2MΩ which still seems a low value in most cases) but the regs are to be applicable to all installations. Very high humidity for example can have dramatic effects on the IR reading. As Note 2) says IR values are usually mich higher than 1M. I would also be worried if the readings were dramatically different for each phase.

The test isn't designed to prove the insulation isn't damaged or isn't perfect. It simply tells you that a) the conductors are not short circuited and b) that the insullation can withstand 1000V with a reasonable current leakage. It's still a critical test though in my opinion.

Note that per American NETA ATS the minumum value is 100MΩ (seems a bit high) but with a note saying that it can be lower due to external factors.

-------------------------
Everyone loves a fireman - but hates the fire inspector.
 22 May 2011 04:17 PM
User is offline View Users Profile Print this message



MBE

Posts: 217
Joined: 13 June 2006

Poor Bat! obviously a victim of incorrectly identfied conductors. He or she could be forgiven for mistaking that live conductor for an earthing conductor. Pitty electricity isnt forgiving.
 22 May 2011 04:26 PM
User is offline View Users Profile Print this message


Avatar for sparkingchip.
sparkingchip

Posts: 5973
Joined: 18 January 2003

Seeing the interest in the Bat try this link:

http://81.171.139.151/WAM/doc/...tion/pdf&pageCount=26

I can assure you there is more money in these reports than there is electrical ones!

Andy
 23 May 2011 08:27 PM
User is offline View Users Profile Print this message



Zs

Posts: 2860
Joined: 20 July 2006

KJ,

Going back to your OP question; yes, I do think the insulation resitance test is valid and viable.

But when I finished at college, the IR test was low on my list of priorities because , by then, I had a completely wasted lump of cable in my tester box, which always came up at >500M yet the bits of insulation were crumbling into the tester box daily. I had never seen <500M (Fluke user). Oh for those halogen days.

I test circuits individually and don't often do a global. That's just a Zs quirk based on the thought that paying for a PIR/T not being like spending money on shoes. I like individual readings on a report and not ditto marks.

To this day I run a continuity test between individual circuit conductors before pumping the voltage into them for an IR test. I think it to be a good starting point/yardstick. Danger Will Robinson.

But then you leave college and you start to see the low readings with which we are all familiar. Now that low result tells a big story, especially if you can see the cable leaving the DB to be sound. It tells you that a problem is probably in a junction or ten, and that is where the dangerous problems often propagate.

Is it too low? Well, IMHO, no. 1 million Ohms. But that comes to you from someone who is out there, at the sharp end and not looking for problems. In the words of a guy on the GN3 stuff 'inspect for goodness, test for badness'. How bad is 1 million Ohms when we see so many installations which are performing perfectly at that level?

I comment in my covering letters if the IR is lower than about 10M. But my first recommendation is to replace socket fronts and JBs. If you can see good cables, with a good degree of bendability in them when you hold them then the low IR test is telling you something else.

And as for the nail through a cable. Well I'd expect to see a good IR test and full serviceabilty of the circuit unless I had been called in to look at a problem. I have a picture of a cooker cable impailed with at least three nails. It measured at >500 and was working perfectly, unless of course you were touching a floorboard nail.

I think the IR test to be a valuable indication of the state of the connections much more than of the state of the cable.

Zs
 25 May 2011 11:13 AM
User is offline View Users Profile Print this message


Avatar for kj scott.
kj scott

Posts: 2144
Joined: 02 April 2006

Zs;
It's not that I think the test invalid; more that the regs and GN3 have it wrong.
612.3.2 requires the test to be completed on a whole installation, or whole distribution board, with a minimum of 1.0 MΩ. This is impractical to apply to all but the simplest installation, such as the 2391 test rig; even domestic installations today, have too many compents that will be affected by, or influence, the 500V test.
It is then more practical to test either during construction, for initial, or for PIR, to test individual circuits, for which there are no minimum values provided. The GN3 approach to guidance is equally as simplistic, making statements like 'To perform the test on a complex installationit may need to be subdivided into its component parts'.
The note below table 61, recognises that where equipment may be damaged, or influence the test, the circuit may be tested at 250V, and the test value should be at least 1MΩ. Calculate the insulation resistance of four circuits at 200MΩ, in parallel with one cicuit at 1MΩ.
I also use a continuity test and a 250V test before the 500V, but there is no guidance on this in GN3, it just jumps straight in.
1.0MΩ, is it too low, not for a whole installation, but it is unrealistic for new work and for a single circuit. I agree that it could still be safe, but it is not what you would expect of a healthy 40m final circuit. A single live cable with a nail through it on a dry plaster finish brick wall will often give a reading of somewhere between 4 and 20MΩ, a timer framed plasterboard wall will be considerably higher.
Yes, the insulation test will find trapped cables or shorts in joints; but how unsafe is an earth fault within a suitably earthed enclosure, or a short circuit within a fire resistant enclosure, where the protective device is suitable co-ordinated with the circuit conductors and the earthing arrangements?

-------------------------
http://www.niceic.biz

Edited: 25 May 2011 at 11:20 AM by kj scott
 25 May 2011 12:20 PM
User is offline View Users Profile Print this message


Avatar for timothyboler                                      .
timothyboler

Posts: 230
Joined: 25 July 2008

Originally posted by: kj scott

Zs;

It's not that I think the test invalid; more that the regs and GN3 have it wrong.

612.3.2 requires the test to be completed on a whole installation, or whole distribution board, with a minimum of 1.0 MΩ. This is impractical to apply to all but the simplest installation, such as the 2391 test rig; even domestic installations today, have too many compents that will be affected by, or influence, the 500V test.

Yes but per 612.3.2 it's considered ok to test the main switchboard and each distribution circuit separately. And unplugging equipment and disconnected a few TVSS and sensitive instruments is not that big a deal. I've tested some large main switchboards before (over 20 sections full of TVSS, instumation CTs etc.) in one go with no problem.


It is then more practical to test either during construction, for initial, or for PIR, to test individual circuits, for which there are no minimum values provided.

The minimum value is 1MΩ whether you test the whole or individual circuits.


The GN3 approach to guidance is equally as simplistic, making statements like 'To perform the test on a complex installation it may need to be subdivided into its component parts'.

Fair point, it's a simple guideline, but it seems a perfectly logical strategy to me. What would you otherwise expect from a guidance note... a detailed methodology for every type of installation?


The note below table 61, recognises that where equipment may be damaged, or influence the test, the circuit may be tested at 250V, and the test value should be at least 1MΩ. Calculate the insulation resistance of four circuits at 200MΩ, in parallel with one cicuit at 1MΩ.

I don't understand your point. Yes the total insulation resistance is less than 1MΩ but we don't care about the total. We care about each circuit individually. If I got 1MΩ on the same type of circuit as the 200MΩ I would be worried. GN3 even recognizes that the IR should be much higher than this and doesn't prevent a competent tester to investigate further to see if there's a problem.


I also use a continuity test and a 250V test before the 500V, but there is no guidance on this in GN3, it just jumps straight in.

Why is this required if you've ensured that sensitive equipment has been disconnencted?


1.0MΩ, is it too low, not for a whole installation, but it is unrealistic for new work and for a single circuit. I agree that it could still be safe, but it is not what you would expect of a healthy 40m final circuit. A single live cable with a nail through it on a dry plaster finish brick wall will often give a reading of somewhere between 4 and 20MΩ, a timer framed plasterboard wall will be considerably higher.

Agreed but GN3 recognizes that in the majority of situations 1MΩ is too low. The regs give a minimum value but that doesn't mean that good judgment from an experienced tester shouldn't be used because an installation just meets the requirements. The IR test is not designed to prove perfect insulation


Yes, the insulation test will find trapped cables or shorts in joints; but how unsafe is an earth fault within a suitably earthed enclosure, or a short circuit within a fire resistant enclosure, where the protective device is suitable coordinated with the circuit conductors and the earthing arrangements?

What further tests (or wording in the guidance) would you propose?Regards, Tim
.

-------------------------
Everyone loves a fireman - but hates the fire inspector.
 25 May 2011 01:24 PM
User is offline View Users Profile Print this message


Avatar for kj scott.
kj scott

Posts: 2144
Joined: 02 April 2006

Originally posted by: timothyboler


Yes but per 612.3.2 it's considered ok to test the main switchboard and each distribution circuit separately. And unplugging equipment and disconnected a few TVSS and sensitive instruments is not that big a deal. I've tested some large main switchboards before (over 20 sections full of TVSS, instumation CTs etc.) in one go with no problem.


The level of dismantling becomes inappropriate to the value of the test, I think you are also over simplifying matters here. I have no problem with unplugging items, but to start pulling recessed presence detectors out of ceilings, and disconnecting 12V lights, emergency lights or high bay fittings, I don't think so.

The minimum value is 1MΩ whether you test the whole or individual circuits.


So 1MΩ is an acceptable value for a circuit 40m length, wired in singles in upvc conduit, with class II accessories?

Fair point, it's a simple guideline, but it seems a perfectly logical strategy to me. What would you otherwise expect from a guidance note... a detailed methodology for every type of installation?


I would expect sufficient information to guide someone. Not simplistic advice based on a domestic final circuit.

I don't understand your point. Yes the total insulation resistance is less than 1MΩ but we don't care about the total. We care about each circuit individually. If I got 1MΩ on the same type of circuit as the 200MΩ I would be worried. GN3 even recognizes that the IR should be much higher than this and doesn't prevent a competent tester to investigate further to see if there's a problem.


612.3.2 requires the total resistance to be greater than 1MΩ, therefore any distribution board which has a final circuit with 1MΩ, will not comply.

Why is this required if you've ensured that sensitive equipment has been disconnencted?


Because it is all too easy to miss something.

Agreed but GN3 recognizes that in the majority of situations 1MΩ is too low. The regs give a minimum value but that doesn't mean that good judgment from an experienced tester shouldn't be used because an installation just meets the requirements. The IR test is not designed to prove perfect insulation


GN3 only reconises that circuits that are less than 2MΩ give cause for concern. This is not realistic.

What further tests (or wording in the guidance) would you propose?Regards, Tim
.


I believe that the guidance should recognise real installations and testing practices, give more examples of problems that could be encountered and some solutions, also perhaps a table of common cable typical insulation resistances, in MΩ/1000m for comparison to measured results.

-------------------------
http://www.niceic.biz
 25 May 2011 01:46 PM
User is offline View Users Profile Print this message


Avatar for rocknroll.
rocknroll

Posts: 8838
Joined: 03 October 2005

I think most people dont really know much about insulation resistance and with all the myths around it gets confusing, I think this simple statement made by BICC who know a little bit about cables sums it up for electrical contractors.

The purpose of the IR test is to check for damaged insulation, this can be mechanical damage or damage by heat, (overloaded cables), readings less than 2 Mohm indicate damaged insulation, readings of 2-50 Mohm are indicative of long circuit lengths, moisture and contamination and do not indicate the insulation quality, therefore readings of 2-50 Mohm should not be specified as a fail value.

regards

-------------------------
"Take nothing but a picture,
leave nothing but footprints!"
-------------------------
"Oh! The drama of it all."
-------------------------
"You can throw all the philosophy you like at the problem, but at the end of the day it's just basic electrical theory!"
-------------------------

Edited: 25 May 2011 at 01:58 PM by rocknroll
 25 May 2011 02:00 PM
User is offline View Users Profile Print this message


Avatar for OMS.
OMS

Posts: 19620
Joined: 23 March 2004

LoL -

I guess I won't argue with those nice chaps over at Callenders then funkmeister

OMS

-------------------------
Failure is always an option
 25 May 2011 03:09 PM
User is offline View Users Profile Print this message


Avatar for kj scott.
kj scott

Posts: 2144
Joined: 02 April 2006

Strange that, from a BICC technical guidance, IR 200 republished February 1991, testing of MICC terminations;
'Test on termination, the value should be noted, this test should then be repeated after 24 hours and the test value should exceed 100MΩ'.
This was of course written back in the days of the 15th Edition, when it was less common to see insulation test instruments with a range >100MΩ.
If a single 50 m circuit of T&E, or singles in conduit, or SWA, installed, could not go off the scale of most test instruments, I would be concerned.

-------------------------
http://www.niceic.biz
 25 May 2011 03:29 PM
User is offline View Users Profile Print this message


Avatar for rocknroll.
rocknroll

Posts: 8838
Joined: 03 October 2005

'Test on termination, the value should be noted, this test should then be repeated after 24 hours and the test value should exceed 100MΩ'.


That was a totally different test when men were men and MI was MI, it related to absorbtion by the insulation material, it was quite common to see a low IR on installation due to the insulation absorbing water from the atmosphere or even because the cable was stored in unfavourable conditions, the IR test determined whether the insulation has dried out in use and also gave an indication that the glands were sealed.

If I remember some of the science from the past is that too much absorbtion would render the insulation useless and no amount of heat would bring it back to its normal state, if this was the case after 24 hours then rip it out and start again.

regards

-------------------------
"Take nothing but a picture,
leave nothing but footprints!"
-------------------------
"Oh! The drama of it all."
-------------------------
"You can throw all the philosophy you like at the problem, but at the end of the day it's just basic electrical theory!"
-------------------------
 25 May 2011 06:12 PM
User is offline View Users Profile Print this message


Avatar for kj scott.
kj scott

Posts: 2144
Joined: 02 April 2006

How many jobs to you get 24 hours to wait and do the testing, even in those days.

-------------------------
http://www.niceic.biz
 25 May 2011 06:44 PM
User is offline View Users Profile Print this message


Avatar for OMS.
OMS

Posts: 19620
Joined: 23 March 2004

Originally posted by: kj scott

How many jobs to you get 24 hours to wait and do the testing, even in those days.


depends if the supervising engineer was on the ball or not - I've tested some bits of MICC three times at least (two dead and again following a "soak")

n
Not commonplace though, granted

OMS

-------------------------
Failure is always an option
 25 May 2011 06:52 PM
User is offline View Users Profile Print this message


Avatar for kj scott.
kj scott

Posts: 2144
Joined: 02 April 2006

I seem to remember MICC ends being made off, and terminated at the same time in most instances, even with the most fastidious consultants.

-------------------------
http://www.niceic.biz
 25 May 2011 06:59 PM
User is offline View Users Profile Print this message


Avatar for OMS.
OMS

Posts: 19620
Joined: 23 March 2004

Made off and terminated yes, because you wanted to avoid water ingress where possible - two tests 24hr apart confirmed that - after a live soak they were often tested again (and the voltage suppressors fitted at that point - on light duty inductive systems) - that was often in addition to the first test the spark did before terminating (to gauge how "damp" the magnesium oxide was and how much needed cutting back before terminating

Regards

OMS

-------------------------
Failure is always an option
 25 May 2011 07:21 PM
User is offline View Users Profile Print this message


Avatar for kj scott.
kj scott

Posts: 2144
Joined: 02 April 2006

Originally posted by: OMS

Made off and terminated yes, because you wanted to avoid water ingress where possible - two tests 24hr apart confirmed that - after a live soak they were often tested again (and the voltage suppressors fitted at that point - on light duty inductive systems) - that was often in addition to the first test the spark did before terminating (to gauge how "damp" the magnesium oxide was and how much needed cutting back before terminating



Regards



OMS


According to BICC, the experts, only 100mm needed to be cut back on the majority of cables, LOL.
Brings me to mind of the bucket test at college, used to see if you had removed the oxide from the cores before sealing the pot.

-------------------------
http://www.niceic.biz
 25 May 2011 07:00 PM
User is offline View Users Profile Print this message


Avatar for rocknroll.
rocknroll

Posts: 8838
Joined: 03 October 2005

Originally posted by: kj scott

How many jobs to you get 24 hours to wait and do the testing, even in those days.


Those were the days when you had proper engineers with proper jobs working for proper companies, even to the extent that anyone caught heating a cable with a blowtorch or zapping it a million times at 1000v was out the door with their toolbag in hot pursuit.

Not like todays breed 'fit and run' especially if there is an RCD involved.

regards

-------------------------
"Take nothing but a picture,
leave nothing but footprints!"
-------------------------
"Oh! The drama of it all."
-------------------------
"You can throw all the philosophy you like at the problem, but at the end of the day it's just basic electrical theory!"
-------------------------
 25 May 2011 07:11 PM
User is offline View Users Profile Print this message


Avatar for OMS.
OMS

Posts: 19620
Joined: 23 March 2004

Those were the days when you had proper engineers with proper jobs working for proper companies,

Yup - as a country we were loosing a f**king fortune in comparison to our competitors

even to the extent that anyone caught heating a cable with a blowtorch or zapping it a million times at 1000v was out the door with their toolbag in hot pursuit.

LoL - or even shorting out a long length and sticking it on an injection tester to get some current circulating to dry it out

Not that we ever did that kind of thing of course - it was the rough chaps on bonus who told me about it


Not like todays breed 'fit and run' especially if there is an RCD involved


Ahh - the wire, fire and f**k off school of installation practice

regards

OMS
OMS

-------------------------
Failure is always an option
 25 May 2011 07:23 PM
User is offline View Users Profile Print this message


Avatar for rocknroll.
rocknroll

Posts: 8838
Joined: 03 October 2005

LOL Thats why I mentioned the blowtorch and 1000v zapping we would never have dreamed about doing that sort of thing would we.

regards

-------------------------
"Take nothing but a picture,
leave nothing but footprints!"
-------------------------
"Oh! The drama of it all."
-------------------------
"You can throw all the philosophy you like at the problem, but at the end of the day it's just basic electrical theory!"
-------------------------
 25 May 2011 07:05 PM
User is offline View Users Profile Print this message



Jaymack

Posts: 4635
Joined: 07 April 2004

Originally posted by: rocknroll
and no amount of heat would bring it back to its normal state, if this was the case after 24 hours then rip it out and start again.

Moisture enters due to bad workmanship at the the pots. The trick is to "chase" the moisture out with a blow lamp, it takes patience though.

Regards
 25 May 2011 07:28 PM
User is offline View Users Profile Print this message


Avatar for OMS.
OMS

Posts: 19620
Joined: 23 March 2004

Originally posted by: Jaymack

Originally posted by: rocknroll

and no amount of heat would bring it back to its normal state, if this was the case after 24 hours then rip it out and start again.


Moisture enters due to bad workmanship at the the pots. The trick is to "chase" the moisture out with a blow lamp, it takes patience though.

Regards


Or when it's un-terminated at first fix

OMS

-------------------------
Failure is always an option
 25 May 2011 07:30 PM
User is offline View Users Profile Print this message


Avatar for OMS.
OMS

Posts: 19620
Joined: 23 March 2004

According to BICC, the experts, only 100mm needed to be cut back on the majority of cables, LOL.


Mmmm - in conventional "dry" constructions maybe - I'd be wanting a bit more than that cut off if it was unsealed and left in shall we say slightly more arduous conditions - and I'd want to see the IR tests

regards

OMS

-------------------------
Failure is always an option
 25 May 2011 07:44 PM
User is offline View Users Profile Print this message


Avatar for kj scott.
kj scott

Posts: 2144
Joined: 02 April 2006

I wasn't saying that was an appropriate length to cut back, just that recommended by the experts.

-------------------------
http://www.niceic.biz
 25 May 2011 08:22 PM
User is offline View Users Profile Print this message


Avatar for GeoffBlackwell.
GeoffBlackwell

Posts: 3516
Joined: 18 January 2003

I was teaching a group British Rail apprentices about terminating MI back in the 70's.

We were using bare MI so I decided to show them how to drive moisture out using a blow lamp.

All went well but I noticed that one of the students had gone into the store with his hot MI in hand - I followed him in and found him cooling the MI under the cold tap - I then asked what he was doing - cooling it down he said - I said why did you heat it up in the first place - strange but true.

Regards

Geoff Blackwell
 25 May 2011 08:26 PM
User is offline View Users Profile Print this message


Avatar for kj scott.
kj scott

Posts: 2144
Joined: 02 April 2006

But it was easy to bend.

-------------------------
http://www.niceic.biz
 25 May 2011 08:59 PM
User is offline View Users Profile Print this message



Jaymack

Posts: 4635
Joined: 07 April 2004

Originally posted by: kj scott
But it was easy to bend.

Not if I know metallurgy!

Regards
 25 May 2011 09:01 PM
User is offline View Users Profile Print this message


Avatar for kj scott.
kj scott

Posts: 2144
Joined: 02 April 2006

It appears not.

-------------------------
http://www.niceic.biz
 25 May 2011 09:17 PM
User is offline View Users Profile Print this message


Avatar for kj scott.
kj scott

Posts: 2144
Joined: 02 April 2006

Have you read the metallurgy book yet Jaymack, and discovered the difference between, annealing and tempering? I think you will find that copper will be flexible after heating and dousing in water.

-------------------------
http://www.niceic.biz
 25 May 2011 09:32 PM
User is offline View Users Profile Print this message



Jaymack

Posts: 4635
Joined: 07 April 2004

Originally posted by: kj scott
Have you read the metallurgy book yet Jaymack, and discovered the difference between, annealing and tempering? I think you will find that copper will be flexible after heating and dousing in water.

Read it? I wrote it and forget half of it! I was really thinking of steel but what would be the effects of having a magnesium oxide core for MIMS?

Regards
 26 May 2011 10:08 AM
User is offline View Users Profile Print this message


Avatar for kj scott.
kj scott

Posts: 2144
Joined: 02 April 2006

Originally posted by: Jaymack

Originally posted by: kj scott

Have you read the metallurgy book yet Jaymack, and discovered the difference between, annealing and tempering? I think you will find that copper will be flexible after heating and dousing in water.


Read it? I wrote it and forget half of it! I was really thinking of steel but what would be the effects of having a magnesium oxide core for MIMS?

Regards


The magnesium oxide has excellent properties when subjected to heat, that is why is is a fire resistant cable. It would be similar to the sand senario and the copper pipe.
Of course the process of heating to drive out moisture would not be as extreme as annealing.


-------------------------
http://www.niceic.biz
 26 May 2011 10:40 AM
User is offline View Users Profile Print this message


Avatar for dg66.
dg66

Posts: 1670
Joined: 11 January 2008

Stainless steel MI was fun

-------------------------
Regards

Dave(not Cockburn)
 26 May 2011 12:26 PM
User is offline View Users Profile Print this message


Avatar for timothyboler                                      .
timothyboler

Posts: 230
Joined: 25 July 2008

Originally posted by: kj scott

The note below table 61, recognises that where equipment may be damaged, or influence the test, the circuit may be tested at 250V, and the test value should be at least 1MΩ. Calculate the insulation resistance of four circuits at 200MΩ, in parallel with one cicuit at 1MΩ.

posted by: timothyboler

I don't understand your point. Yes the total insulation resistance is less than 1MΩ but we don't care about the total. We care about each circuit individually. If I got 1MΩ on the same type of circuit as the 200MΩ I would be worried. GN3 even recognizes that the IR should be much higher than this and doesn't prevent a competent tester to investigate further to see if there's a problem.


posted by: kj scott

612.3.2 requires the total resistance to be greater than 1MΩ, therefore any distribution board which has a final circuit with 1MΩ, will not comply.

posted by: timothyboler

So are you saying that my main switchboard with 20 outgoing distribution circuits measured separately with an IR value of 19.9MΩ each would fail the reg. requirement because the total is less than 1MΩ?

Regards, Tim


-------------------------
Everyone loves a fireman - but hates the fire inspector.
 26 May 2011 01:23 PM
User is offline View Users Profile Print this message


Avatar for GeoffBlackwell.
GeoffBlackwell

Posts: 3516
Joined: 18 January 2003

posted by: timothyboler

So are you saying that my main switchboard with 20 outgoing distribution circuits measured separately with an IR value of 19.9MΩ each would fail the reg. requirement because the total is less than 1MΩ?


Hurray you have finally got there .

Contrary to popular myth - the insulation resistance tests required by 612.3 are whole installation or, at the very least, large section tests - they are not and never have been single circuit tests.

I regularly ask students on the 17th Edition courses I teach whether 11 circuits each tested separately at 10 MOhms is a pass or fail of 612.3.

I bet them £20 - almost all say pass - I could be rich if I actually made them pay .

A test of this type was first suggested in 1882 (Firsts Edition). In those days they were concerned about leakage.

A note in the First Edition
The value of frequently testing the wires cannot be too strongly urged. It is an operation , skill in which is easily acquired and applied. The escape of electricity cannot be detected by the sense of smell, as can gas, but it can be detected by apparatus far more certain and delicate. Leakage not only means waste, but in the presence of moisture it means destruction of the conductor and its insulating covering, by electric action.


The 1937 Electricity Supply Regulations (Statutory) had a rule that said you must not leak more than 1/10000 part of the maximum supply current to earth. This had to be dropped because of supply line filters.

Now you might say that you actually want a single circuit test because leakage is not usually the problem -but damage during installation or in use often is. Well tell JPEL/64 because they have failed to take this on board since the year dot.

You are free to do whatever tests you like within the bounds of safety so you can do single circuit tests if you want - but these do not address 612.3.

Regards

Geoff Blackwell
 26 May 2011 01:39 PM
User is offline View Users Profile Print this message


Avatar for timothyboler                                      .
timothyboler

Posts: 230
Joined: 25 July 2008

I guess I would have owed you £20 as well .

But I might be temped to make a defense that the sentence construction of 612.3.2 does not imply that the IR test applies to the whole....

612.3.2

"...shall be considered satisfactory if the main switchboard AND each distribution circuit tested SEPARATELY, with all ITS final circuits connected...."

Does this not imply that the test of the main switchboard can be seperate to the distribution circuits? The "ITS" assumes the connected final circuits applies to the separate distribution circuit only?

It's a long shot to save me £20

-------------------------
Everyone loves a fireman - but hates the fire inspector.
 26 May 2011 01:59 PM
User is offline View Users Profile Print this message


Avatar for GeoffBlackwell.
GeoffBlackwell

Posts: 3516
Joined: 18 January 2003

with all ITS final circuits connected
.

This is your down fall .

This is telling you to test the circuits as a group - which is exactly what you want if you are testing for leakage from the installation (remember the First edition ).

You are allowed to sub divide large installations because, given that circuit insulation resistances are in parallel, the more cable you have the lower the reading.

This sub division is now based on distribution boards - it used to be based on the number of outlets (50 or less). The change only occurred in 1991 when the 16th Edition came into force.

Regards

Geoff Blackwell
 26 May 2011 02:43 PM
User is offline View Users Profile Print this message


Avatar for kj scott.
kj scott

Posts: 2144
Joined: 02 April 2006

The summation of parallel circuit insulation resistance is taught simply in 2391, whereas the real relationship between individual circuit conductors, other circuit conductors in parallel and multiple earth paths is far more complex.
Also how do you calculate the combined resistance of circuits that exceed the instrument scale reading?
Try it on the next job, measure the individual values, calculate a sum total and measure the whole DB with all loads disconnected, all switches on and all lamps removed.
And of course all equipment disconnected that could be damaged or influence the test result.
Then compare the two results.
Answers on a post card to;
IEE GN3 revision committee working party.

-------------------------
http://www.niceic.biz
 20 May 2011 09:21 PM
User is offline View Users Profile Print this message



mikejumper

Posts: 1706
Joined: 14 December 2006

Originally posted by: kj scott
What insulation resistance value would you expect to see, L-E and L-N on a single circuit wired in T&E buried in a plastered brick wall, when the cable has a nail driven directly into the line conductor only?

Depends on how damp the wall is and whether there are any recessed steel socket or switch boxes, or copper pipes running through it.
 20 May 2011 09:30 PM
User is offline View Users Profile Print this message


Avatar for kj scott.
kj scott

Posts: 2144
Joined: 02 April 2006

Ok; dry wall as in, dry plaster, recessed metal box 300mm away; copper pipework to radiator on same wall within 1.5m.

-------------------------
http://www.niceic.biz
 20 May 2011 09:31 PM
User is offline View Users Profile Print this message


Avatar for sparkingchip.
sparkingchip

Posts: 5973
Joined: 18 January 2003

Another avatar showing something you would not find with a tester, you have to use your eyes!!

Andy

Spot something else about this picture you gold standard testers!
 20 May 2011 09:38 PM
User is offline View Users Profile Print this message


Avatar for slittle.
slittle

Posts: 3487
Joined: 22 November 2007

It's one of my little friends :-(

Found plenty of them over the last couple of years and have photos of most.

The best was between 2 phases on a 200 amp mainswitch. I reckon death was pretty quick

Are they VIR covered tails into a cavity Andy ??, it's hard to tell this time on a friday

Stu
 20 May 2011 09:52 PM
User is offline View Users Profile Print this message


Avatar for kj scott.
kj scott

Posts: 2144
Joined: 02 April 2006

Originally posted by: sparkingchip

Another avatar showing something you would not find with a tester, you have to use your eyes!!



Andy



Spot something else about this picture you gold standard testers!


The rodent was incompetent, has not received adequate training, and was working unsupervised, had it lived it would have made a good tester.

-------------------------
http://www.niceic.biz
 20 May 2011 09:55 PM
User is offline View Users Profile Print this message


Avatar for slittle.
slittle

Posts: 3487
Joined: 22 November 2007

Originally posted by: kj scott

Originally posted by: sparkingchip



Another avatar showing something you would not find with a tester, you have to use your eyes!!







Andy







Spot something else about this picture you gold standard testers!




The rodent was incompetent, has not received adequate training, and was working unsupervised, had it lived it would have made a good tester.


Clearly working without appropriate PPE as well, and no risk or method statement apparent.

I'm looking forward to late June / early July when we start the harvest shakedown to see what the little devils have eaten over the winter.

Stu
 20 May 2011 09:44 PM
User is offline View Users Profile Print this message


Avatar for sparkingchip.
sparkingchip

Posts: 5973
Joined: 18 January 2003

they are tails going into the wall and down the cavity to the meter box, so there is a self evident item to be coded by the gold testers!

Andy
 20 May 2011 09:47 PM
User is offline View Users Profile Print this message


Avatar for JonSteward.
JonSteward

Posts: 567
Joined: 04 December 2007

Died of old age, code 5
 20 May 2011 09:54 PM
User is offline View Users Profile Print this message


Avatar for sparkingchip.
sparkingchip

Posts: 5973
Joined: 18 January 2003

Much simpler, both tails are red!

This poor fellow is dead, possibly old age, left hanging on some tails for eternity.

Andy

photo taken today
 20 May 2011 09:58 PM
User is offline View Users Profile Print this message


Avatar for kj scott.
kj scott

Posts: 2144
Joined: 02 April 2006

That's what killed him then, wasn't sure which one was neutral, took a chance and lost.
I would employ squirrells every time, far more proficient and cost less, they work to keep off the barbeque.

-------------------------
http://www.niceic.biz
 20 May 2011 10:00 PM
User is offline View Users Profile Print this message


Avatar for kj scott.
kj scott

Posts: 2144
Joined: 02 April 2006

Originally posted by: sparkingchip

Much simpler, both tails are red!



This poor fellow is dead, possibly old age, left hanging on some tails for eternity.



Andy



photo taken today


Is that a bat Andy?

-------------------------
http://www.niceic.biz
 20 May 2011 10:03 PM
User is offline View Users Profile Print this message


Avatar for sparkingchip.
sparkingchip

Posts: 5973
Joined: 18 January 2003

long eared and dead.
 20 May 2011 10:06 PM
User is offline View Users Profile Print this message


Avatar for kj scott.
kj scott

Posts: 2144
Joined: 02 April 2006

Protected species then; perhaps the distributor is liable?

-------------------------
http://www.niceic.biz
 20 May 2011 10:06 PM
User is offline View Users Profile Print this message


Avatar for slittle.
slittle

Posts: 3487
Joined: 22 November 2007

Found a live albeit sleepy one in an old TPN switch late last year on one of the farms.

We decided the best option (for the bat) was to isolate the switch from elsewhere and return in a few days. Thankfully he had moved on.


Stu
 20 May 2011 10:22 PM
User is offline View Users Profile Print this message


Avatar for sparkingchip.
sparkingchip

Posts: 5973
Joined: 18 January 2003

Animals and birds do just die, though usually something comes along and eats the corpse. Nature isn't all sweetness and light. I was admiring a Sparrow hawk on our back lawn, then realised it was eating one of the Blackbirds that are nesting in the garden. It still does amaze me how much insulation can be stripped by vermin from cables without killing themselves, and leaving damage not detectable by testing.

Andy
 20 May 2011 10:26 PM
User is offline View Users Profile Print this message


Avatar for kj scott.
kj scott

Posts: 2144
Joined: 02 April 2006

That's 'cos the mice, rats, and squirrels go to training school, where they learn suitable safe working practices.

-------------------------
http://www.niceic.biz
 20 May 2011 10:44 PM
User is offline View Users Profile Print this message



prtadko

Posts: 299
Joined: 09 December 2010

Originally posted by: kj scott

That's 'cos the mice, rats, and squirrels go to training school, where they learn suitable safe working practices.


Thats not a course you run is it KJ?
 20 May 2011 10:55 PM
User is offline View Users Profile Print this message


Avatar for kj scott.
kj scott

Posts: 2144
Joined: 02 April 2006

Over-run with applicants for that one.

-------------------------
http://www.niceic.biz
 20 May 2011 10:54 PM
User is offline View Users Profile Print this message


Avatar for Maxpower.
Maxpower

Posts: 349
Joined: 15 February 2008

Love your avatar, that is one very weird film, still haven't got over it yet and I saw it over 25 years ago.

Ah yes, Eraserhead, haunted me for days and weeks after watching it, then went back to see it again, took yet more drugs, still made no difference... still hum "In Heaven, everything is fine" occasionally...what is it about that film? Oooh, and the radiators, still expect a finned cast iron radiator to hiss at me, and can't eat roast chicken any more...

Sorry, I digress...been meaning to ask: how long are you susceptible to catching Weil's (sp?) disease after rat droppings dry out? Went into a loft the other day, fibreglass insulation was mostly black instead of yellow, closed hatch, told owner to call me when rats, mice & bats had gone and loft fit for human entry.
 21 May 2011 12:03 PM
User is offline View Users Profile Print this message


Avatar for sparkingchip.
sparkingchip

Posts: 5973
Joined: 18 January 2003

Insulation testing has to be the most imprecise area of electrical testing giving the most doubtful results.

As seen above rats can completely strip the insulation from cables, yet they still pass a test. On the other hand part of the installation may be switched off or loads may be in or out of circuit.

On anything other than new installations being tested for the first time a insulation test can only be really be considered a guide to condition at that precise moment in time.

Andy
 25 May 2011 11:38 PM
User is offline View Users Profile Print this message


Avatar for sparkingchip.
sparkingchip

Posts: 5973
Joined: 18 January 2003

Back in those days of old I worked with plumbers who bent 1 1/2" copper pipe by filling it with sand, annealing it then hand bending it, folks have lost the skills of working with copper.

Andy

There is a section of MI in the cardboard box that is the NAPIT museum that survived a fire in Hull Docks, you cannot doubt it's ability to survive heating.

Edited: 26 May 2011 at 08:13 AM by sparkingchip
 26 May 2011 08:57 AM
User is offline View Users Profile Print this message



Jaymack

Posts: 4635
Joined: 07 April 2004

Originally posted by: sparkingchip
Back in those days of old I worked with plumbers who bent 1 1/2" copper pipe by filling it with sand, annealing it then hand bending it, folks have lost the skills of working with copper.

That was a method using for making elbows in large diameter, solid steel pipes ± 3ft. O.D., in the Scottish tube industry up to the 60's, they would plug both ends with wooden bungs after ramming the sand, heat and bend, then machine the ends ready for welding.

Plumbers would probably use an inside spring for their bending nowadays, as with plastic tubing.

Regards
 26 May 2011 01:35 PM
User is online View Users Profile Print this message



AJJewsbury

Posts: 11374
Joined: 13 August 2003

Can't we just test individual circuits and combine the results (resistances in parallel) before comparing with 1M Ohm? Wouldn't that comply? It only says that the 'insulation resistance shall be...', it doesn't say we have to measure it in a single operation. We could even have an extra box on the top of the 'schedule of test results' for the "overall insulation resistance".
- Andy.
 26 May 2011 02:21 PM
User is online View Users Profile Print this message



AJJewsbury

Posts: 11374
Joined: 13 August 2003

This is your down fall

In which case I call the last sentence of 612.1 to my defence ... "Other methods are not precluded provided they give valid results"
- Andy.
 26 May 2011 02:30 PM
User is offline View Users Profile Print this message


Avatar for GeoffBlackwell.
GeoffBlackwell

Posts: 3516
Joined: 18 January 2003

A single circuit test does not given a valid result. Combining the results from a number of single circuit tests would give a similar result but why do many tests when one will do .

The sentence you have referred to concerns the methods detailed in GN3 which is not actually part of BS 7671 (someone should tell JPEL/64 that )

Now if you want a single circuit test lobby JPEL/64 - tell them that it is now the 21 century and that cables are not all that leaky these days .

Regards

Geoff Blackwell
 26 May 2011 02:47 PM
User is offline View Users Profile Print this message


Avatar for kj scott.
kj scott

Posts: 2144
Joined: 02 April 2006

Originally posted by: AJJewsbury

This is your down fall


In which case I call the last sentence of 612.1 to my defence ... "Other methods are not precluded provided they give valid results"

- Andy.


I have a surplus of crystal balls; aquired from a circus, put out of business by section 740, would they do? I can get more accurate results from them than test equipment to BS EN 61557, and they are regularly checked for accuracy against driveby PIR's.

-------------------------
http://www.niceic.biz
 26 May 2011 02:57 PM
User is offline View Users Profile Print this message


Avatar for davezawadi.
davezawadi

Posts: 2642
Joined: 26 June 2002

I like the 612.3.2 note2, this was obviously written by an eternal optimist, who only ever measured small installations with few outgoing circuits, at least if they were not brand new, and even then he might have a few problems.

-------------------------
David
CEng etc, don't ask, its a result not a question!
 26 May 2011 03:13 PM
User is offline View Users Profile Print this message


Avatar for GeoffBlackwell.
GeoffBlackwell

Posts: 3516
Joined: 18 January 2003

Insulation resistance testing when applied to electrical installations is problematic on new works and on periodic testing. The trouble is that there are many sources of error.

Installations of any size present a particular difficulties as the test required is for overall leakage and, as K J Scott has pointed out, this can only be done directly by testing as a group. Combining single circuit results may not be valid as KJ Scott has indicated.

Consider the following:
1) Modern cabling does not generally 'leak' to anything like the degree that older cables did;
2) Modern cabling does not generally deteriorate provided it is used within its specified limits;
3) An installation is far more lightly to 'leak' current via supply line filters and the like. This is deliberate, it is not a defect unless limits are exceeded;
4) Many items of equipment in common use within installations can cause false results;
5) etc, etc - feel free to add to the list .

So is it time to reconsider this test - answers on a postcard please .

Regards

Geoff Blackwell
 26 May 2011 02:38 PM
User is online View Users Profile Print this message



AJJewsbury

Posts: 11374
Joined: 13 August 2003

but why do many tests when one will do

Just 'cos RCCB and RCBOs mess up the test!
- Andy.
 26 May 2011 03:27 PM
User is online View Users Profile Print this message



AJJewsbury

Posts: 11374
Joined: 13 August 2003

Also how do you calculate the combined resistance of circuits that exceed the instrument scale reading?

I'd guess at using the maximum scale as a "erring on the side of safety". e.g. if I had a 48-way DB and got >299 Mohms on each individual circuit, I should be relatively confident that the overall insulation resistance won't be below 1/(48/299M) = 6.22 M ohms - perhaps not pretty, but well within BS 7671 requirements.
- Andy.
 26 May 2011 04:05 PM
User is offline View Users Profile Print this message


Avatar for kj scott.
kj scott

Posts: 2144
Joined: 02 April 2006

Andy,
Your calculation has only considered a single relationship between the conductors, agaist an assumed value.
What about the relationship betwen circuit 1 and all parallel paths when combined with circuits 2-48 and circuit 2 to all parallel paths when combined to circuits 1-48, ................. infinitely.

This is the simplistic idea portrayed by the 2391 exam questions, which serves to mislead.

Taking your example and using one parallel earth path, from a base of 500MΩ for one conductor to earth; you could end up with less than 2.6MΩ

-------------------------
http://www.niceic.biz

Edited: 26 May 2011 at 04:20 PM by kj scott
 26 May 2011 04:20 PM
User is online View Users Profile Print this message



AJJewsbury

Posts: 11374
Joined: 13 August 2003

Andy,
Your calculation has only considered a single relationship between the conductors, agaist an assumed value.
What about the relationship betwen circuit 1 and all parallel paths when combined with circuits 2-48 and circuit 2 to all parallel paths when combined to circuits 1-48, ................. infinitely.

Fair point - my usual method is to disconnect the circuits L & N, but leave the c.p.c. connected (or connect c.pc. & test before connecting L & N) - which I had assumed but hadn't made clear! (oops).

If we say that c.p.c.s should remain connected when testing individual circuits, parallel earth paths (e.g. fault to an extraneous-conductive-part or exposed-conductive-part of another circuit) should be accounted for. I agree we wouldn't necessarily spot faults between live conductors of different circuits - but then a 'full board' test wouldn't spot half of them anyway (L-L and N-N ones) and I'm not sure BS 7671 requires us to take account of faults between different circuits (how would you co-ordinate conductor size and protective device when a conductor from any other circuit could complete the fault path?)

- Andy.
 26 May 2011 04:27 PM
User is offline View Users Profile Print this message


Avatar for kj scott.
kj scott

Posts: 2144
Joined: 02 April 2006

I was not considering fault paths, just parallel insulation resistances. In an attempt to demonstrate that a calculation of sum insulation resistance is unrealistic. As is the measurement of a whole installation or distribution board, in the majority of cases.
We could however compare measured values of insulation resistance against known single circuit conductor characteristics, for types of cable, taken from manufacturer's data, in a similar way we consider conductance, R1, RN, R2 values.

-------------------------
http://www.niceic.biz
 26 May 2011 04:34 PM
User is online View Users Profile Print this message



AJJewsbury

Posts: 11374
Joined: 13 August 2003

Taking your example and using one parallel earth path, from a base of 500MΩ for one conductor to earth; you could end up with less than 2.6MΩ

I'm not quite following that - could you spell it out?

- Andy.
 26 May 2011 04:56 PM
User is offline View Users Profile Print this message


Avatar for kj scott.
kj scott

Posts: 2144
Joined: 02 April 2006

Ok; take one final circuit, in steel conduit
L-E=500MΩ, N-E=500MΩ
L-Conduit =500MΩ N-Conduit =500MΩ
Parallel resistance L+N - E =250MΩ
Parallel resistance L+N - conduit 250MΩ
Sum of the parallel resistances become a further parallel path since conduit and cpc are common= 125MΩ
Now divide into 1, multiply by 48 and divide the total into 1. = 2.6MΩ, without considering the multiple relationships between each final circuit conductor and possibly further parallel earth paths.

Now consider that we don't even know that 500MΩ is a real value, its just off the scale of our test instrument, so we have no chance.

-------------------------
http://www.niceic.biz
 26 May 2011 07:05 PM
User is offline View Users Profile Print this message



Jaymack

Posts: 4635
Joined: 07 April 2004

Originally posted by: kj scott
Parallel resistance L+N - E =250MΩ
Parallel resistance L+N - conduit 250MΩ

Since the conduit and Earth are at the same potential, isn't this just the same resistance measured twice?

Regards
 26 May 2011 07:17 PM
User is offline View Users Profile Print this message


Avatar for OMS.
OMS

Posts: 19620
Joined: 23 March 2004

I'll leave Keith to answer shall I

regards

OMS

-------------------------
Failure is always an option
 26 May 2011 07:38 PM
User is offline View Users Profile Print this message


Avatar for kj scott.
kj scott

Posts: 2144
Joined: 02 April 2006

Originally posted by: Jaymack

Originally posted by: kj scott

Parallel resistance L+N - E =250MΩ

Parallel resistance L+N - conduit 250MΩ


Since the conduit and Earth are at the same potential, isn't this just the same resistance measured twice?

Regards


Correct, at the same potential so having a parallel relationship with the live conductors. Each conductors insulation resistance is in parallel with both the conduit and the cpc, which are common. Its not real, just for example; I wouldn't want to try and calculate an actual installation, as there are too many unknowns.

-------------------------
http://www.niceic.biz
 26 May 2011 08:08 PM
User is offline View Users Profile Print this message



Jaymack

Posts: 4635
Joined: 07 April 2004

Originally posted by: kj scott
Originally posted by: Jaymack
Originally posted by: kj scott
Parallel resistance L+N - E =250MΩ
Parallel resistance L+N - conduit 250MΩ

Since the conduit and Earth are at the same potential, isn't this just the same resistance measured twice?

Correct, at the same potential so having a parallel relationship with the live conductors. Each conductors insulation resistance is in parallel with both the conduit and the cpc, which are common. Its not real, just for example; I wouldn't want to try and calculate an actual installation, as there are too many unknowns.

Why divide the same reading? i.e. "Sum of the parallel resistances become a further parallel path since conduit and cpc are common= 125MΩ"

Regards
 27 May 2011 07:50 AM
User is offline View Users Profile Print this message


Avatar for ebee.
ebee

Posts: 5681
Joined: 02 December 2004

Has anybody ever checked on real installations (new or used) that the sum of parallel circuits works out near to calculated values of adding them together?

On the ones I've checked they usually seem not to, it's not like connecting a few chosen resistors together to obtain a result.

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

Knotted cables cause Lumpy Lektrik
 27 May 2011 01:43 PM
User is offline View Users Profile Print this message


Avatar for kj scott.
kj scott

Posts: 2144
Joined: 02 April 2006

Originally posted by: ebee

Has anybody ever checked on real installations (new or used) that the sum of parallel circuits works out near to calculated values of adding them together?



On the ones I've checked they usually seem not to, it's not like connecting a few chosen resistors together to obtain a result.


This is the point that I am trying to make, 612.3 requires whole installation or whole distribution boards to be >1MΩ. Individual values can not be used to calculate the sum of the circuits of an installation, as could resistors on a PCB. Any single circuit which has 2MΩ, or even 1MΩ quickly falls to an unacceptable level when considered in parallel with the rest of the installation.
Measurement of individual circuits is more practical, but we have no reference point, for the results to be compared with.
Additionally, the values that you calculate are based on test instrument range limits, not actual values, also how would you establish the extent of parallel paths.

-------------------------
http://www.niceic.biz
 27 May 2011 01:33 PM
User is offline View Users Profile Print this message


Avatar for kj scott.
kj scott

Posts: 2144
Joined: 02 April 2006

Originally posted by: Jaymack

Why divide the same reading? i.e. "Sum of the parallel resistances become a further parallel path since conduit and cpc are common= 125MΩ"

Regards


Because in the first calculation they were dealt with as separate values, however since they are in parallel to both the conduit and the cpc their resistances will be parallel.

-------------------------
http://www.niceic.biz
 27 May 2011 02:11 PM
User is offline View Users Profile Print this message



Jaymack

Posts: 4635
Joined: 07 April 2004

Originally posted by: kj scott
Originally posted by: Jaymack
Why divide the same reading? i.e. "Sum of the parallel resistances become a further parallel path since conduit and cpc are common= 125MΩ"
Regards

Because in the first calculation they were dealt with as separate values, however since they are in parallel to both the conduit and the cpc their resistances will be parallel.

They are both common nodes, the conduit and CPC are common and the shorted Lives are common, how are they in parallel?

Regards
 27 May 2011 02:42 PM
User is offline View Users Profile Print this message


Avatar for kj scott.
kj scott

Posts: 2144
Joined: 02 April 2006

You are thinking in terms of resistors, the live conductors are run inside the conduit and so in parallel in relation to it, the same is true of the cpc. Although the cpc is common to the conduit it is also parallel to the live conductors, hence has a relationship that has to be calculated twice. Draw it it will make sense, and remember it is only a theoretical exercise.

-------------------------
http://www.niceic.biz
 27 May 2011 03:10 PM
User is offline View Users Profile Print this message



Jaymack

Posts: 4635
Joined: 07 April 2004

Originally posted by: kj scott
You are thinking in terms of resistors, the live conductors are run inside the conduit and so in parallel in relation to it, the same is true of the cpc. Although the cpc is common to the conduit it is also parallel to the live conductors, hence has a relationship that has to be calculated twice. Draw it it will make sense, and remember it is only a theoretical exercise.

I'm thinking of the resistance measurements (1) between the lives connected and the CPC; and (2) between the same lives connected and Earth, (the CPC and Earth being connected together). This will give 2 readings that are largely measuring the same I.R., I don't see that they are in parallel to the extent that the I.R. readings can be halved as stated, it is a common I.R. reading measured twice.

Regards
 27 May 2011 03:40 PM
User is offline View Users Profile Print this message


Avatar for kj scott.
kj scott

Posts: 2144
Joined: 02 April 2006

If you go back to my original example, the initial resistance given was for one conductor, to earth and one conductor to the conduit. These are then calculated in parallel, against two common paths, then considered as a multiple of fortyeight times.
As I also said it is a theoretical example of the effects of parallel resistance, in reality the extent to which the parallel paths exist would be unknown and certainly not uniform; it has been used to demonstrate how a value of 500MΩ could rapidly diminish.

-------------------------
http://www.niceic.biz
 27 May 2011 04:19 PM
User is offline View Users Profile Print this message



Jaymack

Posts: 4635
Joined: 07 April 2004

Originally posted by: kj scott
If you go back to my original example, the initial resistance given was for one conductor, to earth and one conductor to the conduit. These are then calculated in parallel, against two common paths, then considered as a multiple of fortyeight times.
As I also said it is a theoretical example of the effects of parallel resistance, in reality the extent to which the parallel paths exist would be unknown and certainly not uniform; it has been used to demonstrate how a value of 500MΩ could rapidly diminish.

This is your statement that I was referring to: -

Parallel resistance L+N - E =250MΩ
Parallel resistance L+N - conduit 250MΩ
Sum of the parallel resistances become a further parallel path since conduit and cpc are common= 125MΩ


Why would it be a further parallel path, if the I.R. measurements were taken across common rails?

Regards
 27 May 2011 04:50 PM
User is offline View Users Profile Print this message


Avatar for kj scott.
kj scott

Posts: 2144
Joined: 02 April 2006

Because the cpc and the conduit are both parallel to the insulation of the live conductors. As I have said you are thinking in terms of resistors on a pcb, not that of an electrical installation.

-------------------------
http://www.niceic.biz
 27 May 2011 05:42 PM
User is offline View Users Profile Print this message



Jaymack

Posts: 4635
Joined: 07 April 2004

Originally posted by: kj scott
Because the cpc and the conduit are both parallel to the insulation of the live conductors. As I have said you are thinking in terms of resistors on a pcb, not that of an electrical installation.

It matters not a jot whether they are all parallel in this case, nodes, points of connection and rails etc., all have the same principles of calculations including those for resistances in parallel. Your specific statement on halving the I.R. is the one being questioned here, the example given is an insulation measurement of common Live conductors to a common CPC/Earth.

Regards

Regards
 27 May 2011 09:54 PM
User is offline View Users Profile Print this message


Avatar for kj scott.
kj scott

Posts: 2144
Joined: 02 April 2006

Ok lets do this slowly;
If I have a single conductor in a steel conduit, which has an insulation resistance of 500M ohms and a length of 100m.

Then I add another conductor in parallel of the same size with the same insulation resistance, what would be the total insulation resistance?

-------------------------
http://www.niceic.biz
 27 May 2011 10:22 PM
User is offline View Users Profile Print this message


Avatar for ebee.
ebee

Posts: 5681
Joined: 02 December 2004

One might expect 250M ohms

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

Knotted cables cause Lumpy Lektrik
 27 May 2011 10:54 PM
User is offline View Users Profile Print this message


Avatar for Jobbo.
Jobbo

Posts: 912
Joined: 08 July 2010

Interesting thread

Would consideration be given for the current IR value of an installation, when adding an additional final circuit to an existing installation, with many final circuits? I think not. Additional final circuits would be tested on an individual basis not part of a global test, yet it will have influence on the complete installation, possibly dropping the IR below the 1M value. As Keith has stated, a figure should be given in BS7671 for a single circuit, not just for distribution circuits

Jobbo
 27 May 2011 11:37 PM
User is offline View Users Profile Print this message


Avatar for kj scott.
kj scott

Posts: 2144
Joined: 02 April 2006

Originally posted by: Jobbo

Interesting thread



Would consideration be given for the current IR value of an installation, when adding an additional final circuit to an existing installation, with many final circuits? I think not. Additional final circuits would be tested on an individual basis not part of a global test, yet it will have influence on the complete installation, possibly dropping the IR below the 1M value. As Keith has stated, a figure should be given in BS7671 for a single circuit, not just for distribution circuits



Jobbo


Come on jobbo; just phone me.
Sorry just read the whole post, it is Friday.

-------------------------
http://www.niceic.biz
 28 May 2011 06:17 AM
User is offline View Users Profile Print this message


Avatar for GeoffBlackwell.
GeoffBlackwell

Posts: 3516
Joined: 18 January 2003

Originally posted by: Jobbo

Interesting thread

Would consideration be given for the current IR value of an installation, when adding an additional final circuit to an existing installation, with many final circuits? I think not. Additional final circuits would be tested on an individual basis not part of a global test, yet it will have influence on the complete installation, possibly dropping the IR below the 1M value. As Keith has stated, a figure should be given in BS7671 for a single circuit, not just for distribution circuits

Jobbo


As I said earlier in this topic:

Contrary to popular myth - the insulation resistance tests required by 612.3 are whole installation or, at the very least, large section tests - they are not and never have been single circuit tests.


Now you might say that you actually want a single circuit test because leakage is not usually the problem -but damage during installation or in use often is. Well tell JPEL/64 because they have failed to take this on board since the year dot.


So there you have it - the intended purpose of the test is not what many of you seem to assume that is - it was ever thus .

Now if you feel that what is really need in the 21st century is a single circuit test - Well tell JPEL/64 because they have failed to take this on board since the year dot.

You could then expect limit values to be in 10's or 100's of MOhms per new circuit.

Regards

Geoff Blackwell
 28 May 2011 06:47 AM
User is offline View Users Profile Print this message


Avatar for ebee.
ebee

Posts: 5681
Joined: 02 December 2004

Well a figure of X ohms per installation does seem a bit silly as then you apply it to whole installations irrespective of whether such installation contains 1, 10, 100 or 1000 final circuits.

I suppose something along the lines of 2M ohm for the whole installation and 10M ohm for any individual circuit for instance sounds a bit more sensible.
Or some such other figures perhaps.

----------------------------------------

Does anyone else find, like I have done a few times, that for instance say 10 circuits leaving all cpcs & bonding connected as one "earth connection" and testing to lives say 10 circuits at 10M ohm each does not produce an overall IR of 1M ohm but probably somewhere between say 5 to 9 M ohm?

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

Knotted cables cause Lumpy Lektrik
 28 May 2011 09:48 AM
User is offline View Users Profile Print this message


Avatar for OMS.
OMS

Posts: 19620
Joined: 23 March 2004

So there you have it - the intended purpose of the test is not what many of you seem to assume that is - it was ever thus


For sure Geoff - 612.3.2 and it's predecessors have never discussed single circuit IR - that was never an issue because the discretion for single circuit testing has always been in 612.3.1.

I guess the problem is people not understanding the relationship between the two issues and assumng that if every circuit presents a "good" IR individually then it will comply with the required test of 612.3.2.

Personally, I never understood why JPEL took out the requirement in the former 713-04-01 for the test to be undertaken before connection to a supply

regards

OMS

-------------------------
Failure is always an option
 28 May 2011 09:50 AM
User is offline View Users Profile Print this message



Jaymack

Posts: 4635
Joined: 07 April 2004

Originally posted by: Jobbo
As Keith has stated, a figure should be given in BS7671 for a single circuit, not just for distribution circuits

Agreed ......... for initial verification, I wouldn't agree for a PIR though, when a minimum acceptable value is given, albeit where a trend is apparent from previous records.

Regards
 28 May 2011 10:02 AM
User is offline View Users Profile Print this message


Avatar for ebee.
ebee

Posts: 5681
Joined: 02 December 2004

I was commoning them thru neutral bars and terminal posts ie no on board RCD to consumer unit and measuring each circuit individually then as a whole .

It don't always give readings somewhere in the parish of what you'd expect if you'd used pure resistors for the excersize.

I calibrate mee own squeeze on a regular basis Geoff

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

Knotted cables cause Lumpy Lektrik
 27 May 2011 11:36 PM
User is offline View Users Profile Print this message


Avatar for kj scott.
kj scott

Posts: 2144
Joined: 02 April 2006

Originally posted by: ebee

One might expect 250M ohms


So first stage; where is jaymack?

-------------------------
http://www.niceic.biz
 28 May 2011 09:39 AM
User is offline View Users Profile Print this message



Jaymack

Posts: 4635
Joined: 07 April 2004

Originally posted by: ebee
One might expect 250M ohms

Agreed, not what my posts are about previous posts though! If you take an I.R. reading between commoned live conductors and commoned Earth/CPC of 250 Mohms, what would be expected for a subsequent reading, using the same points of measurement?.

Regards
 28 May 2011 09:42 AM
User is offline View Users Profile Print this message


Avatar for GeoffBlackwell.
GeoffBlackwell

Posts: 3516
Joined: 18 January 2003

ebee wrote:
Does anyone else find, like I have done a few times, that for instance say 10 circuits leaving all cpcs & bonding connected as one "earth connection" and testing to lives say 10 circuits at 10M ohm each does not produce an overall IR of 1M ohm but probably somewhere between say 5 to 9 M ohm?


Squeeze your croc clips - but be sure to use a calibrated squeeze .

Regards

Geoff Blackwell
 28 May 2011 11:26 AM
User is offline View Users Profile Print this message


Avatar for kj scott.
kj scott

Posts: 2144
Joined: 02 April 2006

Originally posted by: Jaymack

Originally posted by: ebee

One might expect 250M ohms


Agreed, not what my posts are about previous posts though! If you take an I.R. reading between commoned live conductors and commoned Earth/CPC of 250 Mohms, what would be expected for a subsequent reading, using the same points of measurement?.

Regards


Ok, now what would the insulation resistance be if we double the length of the circuit to 200m?

-------------------------
http://www.niceic.biz
 28 May 2011 12:21 PM
User is offline View Users Profile Print this message



Jaymack

Posts: 4635
Joined: 07 April 2004

Originally posted by: kj scott
Originally posted by: Jaymack
Originally posted by: ebee
One might expect 250M ohms

Agreed, not what my posts are about previous posts though! If you take an I.R. reading between commoned live conductors and commoned Earth/CPC of 250 Mohms, what would be expected for a subsequent reading, using the same points of measurement?.
Regards

Ok, now what would the insulation resistance be if we double the length of the circuit to 200m?


........ for you!

Regards
 28 May 2011 11:43 AM
User is offline View Users Profile Print this message


Avatar for sparkingchip.
sparkingchip

Posts: 5973
Joined: 18 January 2003

That could depend on how long I keep my finger on the test button for, and the range on my test meters 200 for the Robin and 1000 for the Megger, real life insulation testing of old installations is only a approximation and does not give a figure that can be figure set in stone, however I would guess it would halve on the initial reading.

Andy
 28 May 2011 11:59 AM
User is offline View Users Profile Print this message


Avatar for kj scott.
kj scott

Posts: 2144
Joined: 02 April 2006

This is only theory sparkingchip, what would really happen is anyones guess

-------------------------
http://www.niceic.biz
 28 May 2011 12:24 PM
User is offline View Users Profile Print this message


Avatar for kj scott.
kj scott

Posts: 2144
Joined: 02 April 2006

Originally posted by: sparkingchip

I would guess it would halve on the initial reading.

Andy


So that would be 125MΩ, does Jaymack agree?

-------------------------
http://www.niceic.biz
 28 May 2011 12:00 PM
User is offline View Users Profile Print this message



ant1uk

Posts: 2376
Joined: 22 June 2010

I was testing an older installation made sure everything was disconnected in the whole house. was getting 500M on L-N and L-E but when I tested N-E it started at 100M then was going up slowly to 500M

I have never seen this before and wondered why?
 28 May 2011 12:14 PM
User is offline View Users Profile Print this message


Avatar for OMS.
OMS

Posts: 19620
Joined: 23 March 2004

Originally posted by: ant1uk

I was testing an older installation made sure everything was disconnected in the whole house. was getting 500M on L-N and L-E but when I tested N-E it started at 100M then was going up slowly to 500M
I have never seen this before and wondered why?


A combination of capacitance and absorbtion in the cable insulation

A steadily rising value is what you want to see - this shows that the insulation is good - if it's damaged or contaminated in any way then the leakage masks the absorbtion and you get a fairly constant (lower) value

OMS

-------------------------
Failure is always an option
 28 May 2011 12:19 PM
User is offline View Users Profile Print this message



ant1uk

Posts: 2376
Joined: 22 June 2010

So do you think its nothing to worry about? it just seemed unusual as the test normally goes straight to 500M or similar. I think there was a flourecent light in the garage which I couldn't access so i can not confirm this. do you think this could be a possible cause?
 28 May 2011 12:23 PM
User is offline View Users Profile Print this message


Avatar for OMS.
OMS

Posts: 19620
Joined: 23 March 2004

Possibly, you could have been charging the PF capacitor. It would have the same effect as you charging a cable.

A steadily rising value isn't usually one to worrry about - the properties of something like older "capathene" insulation compared to "PVC" are different so you will see different effects

Regards

OMS

-------------------------
Failure is always an option
 28 May 2011 12:29 PM
User is offline View Users Profile Print this message



ant1uk

Posts: 2376
Joined: 22 June 2010

Well that may explain it OMS I noticed some parts of the installation were rewired at some point but there was still some old cables in some parts which maybe as you suggested. I thought all was well anyway as there was no straight shorts anywhere.
 29 May 2011 08:05 AM
User is offline View Users Profile Print this message


Avatar for davezawadi.
davezawadi

Posts: 2642
Joined: 26 June 2002

Of course the BS7671 schedule of test results does have a value for each circuit, so keep on testing!

-------------------------
David
CEng etc, don't ask, its a result not a question!
 29 May 2011 08:41 AM
User is offline View Users Profile Print this message



Phoenix151

Posts: 77
Joined: 14 February 2007

Perhaps one of you enlightened chaps can offer an explanation for something i've now seen a few times, perhaps JP has seen similar?

Doing an IR test across the whole DB as GB tells us that we should ; unplug all the portable appliances that can be found, isolate, prove, link phase(s) and neutral, drop neutral link if neutral not through the isolator, test 'poles to earth' as it was called by a old college lecturer who tought me.

I have yet to see one >200, but seen loads that are right down. But the issue I wanted to ask about is that I initially test at 250v and then only if thats above 2 meg do I do it with 500v, I have seen a few occasions where the 250v result has been about 10meg, but the 500v one has been right down, and switching back to 250v still gives a result thats alomst on the floor. First thought is that something electronic has been caught in the cross fire, but these boards supply only sockets (with everything unplugged) and lighting. Nothing has failed to function afterwards and even a few months later no one has come back to me (and this is a site where you get admonished if you accidentally leave a computer unpluged!).

Only thing I can think of is that most times I've seen this, that lighting has been quite well cooked inside the fittings, and I'm guessing the internal insulation of the magnetic chokes on its last legs? anyone got any better theories

(before I go and play with the chokes of some fittings removed from the skip and a blow lamp )
 29 May 2011 10:15 AM
User is offline View Users Profile Print this message



John Peckham

Posts: 7418
Joined: 23 April 2005

Phoenix

Most books and contributions on this forum relate to domestic T&E installations. Large commercial and industrial installations are not scaled up domestic T&E jobs unless a domestic electrician has done the work.

IR results on large boards full of circuits feeding flourescent lighting and building services plant never give >200 readings despite what you might see on test result forms. Electronic ballasts, EL fittings and fixed plant drive down the IR. As I said before if we get a big DB over 1M it's cream cakes all round. You don't need many old ballasts in parallel to screw the IR.

We tested a floor in a large London building yesterday 2 of the DBs were sub 1 ohm neutral to earth on the meter continuity range. A 5 year old installation with floor tracks and home and away boxes feeding the lighting and fan coils units. A few radial circuits to BES plant. As usual pages of defects some of which were there from the original install such as undersized SWA to the DBs from the bus bar tap off boxes, 35mm 4 core fed from 250A MCCB in tap off.

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

http://www.astutetechnicalservices.co.uk/
 29 May 2011 11:43 AM
User is offline View Users Profile Print this message


Avatar for dg66.
dg66

Posts: 1670
Joined: 11 January 2008

Could the 250A MCCBs not be adjusted down?

-------------------------
Regards

Dave(not Cockburn)
 29 May 2011 12:03 PM
User is offline View Users Profile Print this message



Phoenix151

Posts: 77
Joined: 14 February 2007

JP, you've missed the subtleties of me saying that i've never seen a board read >200. I know its unreasonable, I was just pointing that I too have never seen it dispite what the previous tests often say

I'm happy if I see >5meg, and thats normally somewhere around 40% - 50% of the time, most of what I test is 8 / 12 way boards in educational buildings, depending on the age/exact nature of the building it tends to either be 6242Y on basket tray or singles in conduit and trunk

I think I might have to limit myself to a cream cake for every 5th one that passes.... got to watch the waistline

Do you often see poor work from big national contracting names john?, I've seen jobs where grommets were missing in about half the points across the building, and the dual earth computer sockets had both ring final cpcs to one terminal and a flylead across to the other (obviously and apprentice mis-understood when told to connect to both terminals!)

Oh, and I see your 35mm off a 250A breaker and raise you a 35mm feeding an extra 16way tpn final DB (rated at 100 or 125A can't remember) stuffed with 32A 3ph breakers, which had been doubled up in the main panel onto the terminals of the 400A breaker suppling the panel board at the other end of the factory!

A.
 29 May 2011 12:22 PM
User is offline View Users Profile Print this message


Avatar for dg66.
dg66

Posts: 1670
Joined: 11 January 2008

The company i recently worked for seemed to ignore the insulation resistance values i recorded on site,and all the typed up completed schedules of test results all came in with +200MOhm values,and the inspection reports detailing non compliances seemed to get ignored and all certification was signed off ,with no issues,coupled with the fact that some of the personell still employed have limited technical knowledge,no 2391 and are quite happy to fabricate test results.

-------------------------
Regards

Dave(not Cockburn)
 29 May 2011 12:24 PM
User is offline View Users Profile Print this message


Avatar for sparkingchip.
sparkingchip

Posts: 5973
Joined: 18 January 2003

I worked on a big old house with a thirty year old installation updated with new consumer units about three years ago recently. The first unit has a 100mA S-type RCD main switch and feeds the 2nd unit, which in turn feeds a 3rd unit in total 28 circuits, up font of the whole lot is a 60 amp fused main switch which jammed and needed replacing to turn the power back on, insulation test fails as to be expected the leakage is around 6mA over all of the 28 circuits that is just over 214 micro amps each, You could spend several days trying to raise the test result figures, I'd like to see the certificates from the "updating" carried out around three years ago, I bet all the individual circuit readings are good.

Andy
 29 May 2011 06:03 PM
User is offline View Users Profile Print this message



John Peckham

Posts: 7418
Joined: 23 April 2005

DG66

Yes the MCCBs could be tirned down but not by that much. Also would yhou adjust an MCCB at the time of a PIR on a weekend in a London City Bank that trades billions in the week?

Phoenix

Yes I see loads of poor installation work by both big and small contractors. Often the bigger jobs look fine from a 1st impression but closer scrutiny proves other wise.

Andy

I have tried to explore earth leakage measurement vs IR testin on this forum without much sucess. We often use this when we cannot isolate but I would be interested on other people's input.

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

http://www.astutetechnicalservices.co.uk/
 29 May 2011 07:01 PM
User is offline View Users Profile Print this message



Zs

Posts: 2860
Joined: 20 July 2006

This is the modern world.

As already stated in one of this weekend's posts, sorry I can't find it again but it is a page or two back; a full DB IR test, of a system with up to date protection will not give a satisfactory result as one test. DBs have to be broken down these days, at least beyond the RCDs in order to offer anything like a genuine picture.

612.3.2 is clearly written from a conference table and without a thought to the testing of smaller installations. It assumes a chunk of main switch gear and a sub DB. Probably without RCDs and other din rail mounted kit/electronics and so on. I would imagine that more domestic installations are subject to a test and inspect every day.

So it is nonsense. But probably only because of poor wording and forgetting the bigger picture.

ebee and I know how important that test is. And we know that the magic calculation taught at college is not immutable when you have dirty hands and a tester on the go, with your head sticking in the cupboard under the stairs. Which means loads of us know the reality of IR testing. The circuits we are testing are rarely pure and as such the physics behind the IR global calcs isn't going to work. Too many variables when you get away from a desk and start meeting the spiders and dead mice. Same as Zs not really equalling Ze+.....

Yes, it remains valid. And No, 1M is not too low.

Zs
 07 June 2011 08:20 PM
User is offline View Users Profile Print this message


Avatar for kj scott.
kj scott

Posts: 2144
Joined: 02 April 2006

Originally posted by: Zs

This is the modern world.

Yes it is, and we need to change the specification for testing and guidance to suit.

As already stated in one of this weekend's posts, sorry I can't find it again but it is a page or two back; a full DB IR test, of a system with up to date protection will not give a satisfactory result as one test. DBs have to be broken down these days, at least beyond the RCDs in order to offer anything like a genuine picture.

Agreed, but how far is too far, before dismantling becomes a greater problem than the value of the test result?

612.3.2 is clearly written from a conference table and without a thought to the testing of smaller installations. It assumes a chunk of main switch gear and a sub DB. Probably without RCDs and other din rail mounted kit/electronics and so on. I would imagine that more domestic installations are subject to a test and inspect every day.

It would appear that the guidance dates back to previous versions of test methodology, without consideration of the practical aspects or current installation conditions.

So it is nonsense. But probably only because of poor wording and forgetting the bigger picture.

So does it not need a review?

ebee and I know how important that test is. And we know that the magic calculation taught at college is not immutable when you have dirty hands and a tester on the go, with your head sticking in the cupboard under the stairs. Which means loads of us know the reality of IR testing. The circuits we are testing are rarely pure and as such the physics behind the IR global calcs isn't going to work. Too many variables when you get away from a desk and start meeting the spiders and dead mice. Same as Zs not really equalling Ze+.....

So if the test result is unreliable, what value the test?

Yes, it remains valid. And No, 1M is not too low.

1MΩ, is generally too low for a single circuit, unless very long/large/arduous conditions. It is not even realistic for a small installation; it allows serious faults to pass un-noted.
Zs


-------------------------
http://www.niceic.biz
 31 May 2011 12:49 PM
User is online View Users Profile Print this message



AJJewsbury

Posts: 11374
Joined: 13 August 2003

Ok; take one final circuit, in steel conduit
L-E=500MΩ, N-E=500MΩ
L-Conduit =500MΩ N-Conduit =500MΩ
Parallel resistance L+N - E =250MΩ
Parallel resistance L+N - conduit 250MΩ
Sum of the parallel resistances become a further parallel path since conduit and cpc are common= 125M

But wouldn't that be the case whether we were testing an entire DB in one go or circuit-by-circuit?

I suspect it's not as bad as that though. From a 'Laws of physics' point of view, Testing L-N doesn't ignore the c.p.c. insulation entirely during the test. (ignoring the conduit for a moment) you'll have L-c.p.c and then c.p.c-N resistances in series, in parallel with the L-N insulation resistance, already included in the L-N figure shown by the meter.

So if your meter reads 500Mohm L-N, it could mean that L-N is actually 750M with another 750M between L-c.p.c. and 750M between c.p.c.-N. (1/(1/750 + 1/(750+750)))

So some adjustment for parallel paths has been automatically taken into account.

Now divide into 1, multiply by 48 and divide the total into 1. = 2.6MΩ, without considering the multiple relationships between each final circuit conductor and possibly further parallel earth paths.

Now consider that we don't even know that 500MΩ is a real value, its just off the scale of our test instrument, so we have no chance.

I guess it depends on what we're trying to achieve with the test. If to check for faults, there's probably no argument (it'll be go/no-go based on whether the meter shows 0.00 or not). If it's to limit earth leakage (e.g. to show compliance with the old supply regs) then we're only really interested in current flowing from L to earth (or N combined). Insulation between one L wire and another L wire isn't of interest (as it won't contribute to the leakage current), likewise insulation between N and PE (presuming a TN or TT system).

Yes, there's resistances between one circuit's L and another's N/PE, but with most wiring systems that's likely to be small compared with the circuit's own. I guess worst case would be singles in trunking, but even then the insulation of the wiring probably isn't as significant as the connected equipment or specific dirt/damp problems that probably wouldn't be significantly effected by parallel paths.

My gut feel is that summing the results for individual circuits (by resistances in parallel) would give a reasonable indication of the value to be expected from a full board test. Perhaps what we need is some good experimental data though.

- Andy.

(edited to fix quoting)
 02 June 2011 05:23 PM
User is offline View Users Profile Print this message



lyledunn

Posts: 617
Joined: 13 August 2003

So the IR test on a TPN board reveals fairly low results, shall we say 0.03Mohms. What do we do? Report it and let the client worry about it? Pull the board apart with all that that might entail?
Perhaps what we could do, at least as far as PIRs are concerned, is accept that a reading of 1Mohm or more is good but on the other hand a reading of much less than that is not necessarily bad.
It is not valid to compare the test result to Table 61 if equipment is attached and I imagine most would agree that the complete isolation of current using equipment in an existing installation is close to impossible.
If there really was serious concern about low values of IR in fixed installations then 4-pole isolation of distribution boards would have been an essential requirement along with the need to dispense with the common neutral bar.

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

Lyle Dunn
 03 June 2011 01:13 PM
User is offline View Users Profile Print this message


Avatar for OMS.
OMS

Posts: 19620
Joined: 23 March 2004

Originally posted by: lyledunn

So the IR test on a TPN board reveals fairly low results, shall we say 0.03Mohms. What do we do? Report it and let the client worry about it? Pull the board apart with all that that might entail?

Perhaps what we could do, at least as far as PIRs are concerned, is accept that a reading of 1Mohm or more is good but on the other hand a reading of much less than that is not necessarily bad.

It is not valid to compare the test result to Table 61 if equipment is attached and I imagine most would agree that the complete isolation of current using equipment in an existing installation is close to impossible.

If there really was serious concern about low values of IR in fixed installations then 4-pole isolation of distribution boards would have been an essential requirement along with the need to dispense with the common neutral bar.


Well you could easily have a discernible flow of current at 0.03Megohm so I wouldn't be ignoring it. If your simply testing then Code it and move on - if only a 3

Regards

OMS

-------------------------
Failure is always an option
 07 June 2011 11:51 PM
User is offline View Users Profile Print this message



lyledunn

Posts: 617
Joined: 13 August 2003

OMS replied; "Well you could easily have a discernible flow of current at 0.03Megohm so I wouldn't be ignoring it. If your simply testing then Code it and move on - if only a 3"
OMS, I would see little value in such a response, at least from the clients perspective. I do not simply test nor indeed should testing be viewed as some kind of stand-alone procedure in conducting a periodic inspection . A code 3 is a cop-out and should only be arrived at as a last resort. I test on commercial PIRs most days of the week and I always endeavour to carry out IR tests where possible. It is rare that IR levels between NandE on distribution boards (with final ccts attached) are above 2M with many failing to achieve 0.5M. Thus with equipment connected, the Line-E values will register low. The logistics of investigation are difficult and simply batting it back to the client seems more than a tad un-professional. If you feel that an IR value of 0.03M is a problem then you must advise the client, but to do what?Spend hours dismantling the neutral bar, disconnecting equipment and causing all sorts of upset only to find the offender is only the front-end filter of a vsd or the like!

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

Lyle Dunn
 13 June 2011 09:58 AM
User is offline View Users Profile Print this message


Avatar for kj scott.
kj scott

Posts: 2144
Joined: 02 April 2006

Just thought I would bring this topic to the top again after seeing flashtestdummy post below.

-------------------------
http://www.niceic.biz
 13 June 2011 10:17 AM
User is offline View Users Profile Print this message


Avatar for OMS.
OMS

Posts: 19620
Joined: 23 March 2004

Originally posted by: lyledunn

OMS replied; "Well you could easily have a discernible flow of current at 0.03Megohm so I wouldn't be ignoring it. If your simply testing then Code it and move on - if only a 3"

OMS, I would see little value in such a response, at least from the clients perspective. I do not simply test nor indeed should testing be viewed as some kind of stand-alone procedure in conducting a periodic inspection . A code 3 is a cop-out and should only be arrived at as a last resort. I test on commercial PIRs most days of the week and I always endeavour to carry out IR tests where possible. It is rare that IR levels between NandE on distribution boards (with final ccts attached) are above 2M with many failing to achieve 0.5M. Thus with equipment connected, the Line-E values will register low. The logistics of investigation are difficult and simply batting it back to the client seems more than a tad un-professional. If you feel that an IR value of 0.03M is a problem then you must advise the client, but to do what?Spend hours dismantling the neutral bar, disconnecting equipment and causing all sorts of upset only to find the offender is only the front-end filter of a vsd or the like!


Well, your the man undertaking the PIR Lyle and I've no idea what contractual arrangement exists between you and your client.

I was simply pointing out that presented with a reading of 0.03Mohm on a group of circuits it should be further investigated as it clearly doesn't align with BS 7671.

There may be many reasons as to why and full investigation may well require extensive dismantling of the installation - whether that is acceptable to you or your client isn't really an electrical issue is it ?

As for it being unprofessional, well it depends on what you are being asked to do - is it to provide an impartial, unbiased report on the installation or is it help your client limp an installation through to a satisfactory outcome

Regards

OMS

-------------------------
Failure is always an option
 07 June 2011 08:26 PM
User is offline View Users Profile Print this message


Avatar for kj scott.
kj scott

Posts: 2144
Joined: 02 April 2006

Originally posted by: AJJewsbury




I suspect it's not as bad as that though. From a 'Laws of physics' point of view, Testing L-N doesn't ignore the c.p.c. insulation entirely during the test. (ignoring the conduit for a moment) you'll have L-c.p.c and then c.p.c-N resistances in series, in parallel with the L-N insulation resistance, already included in the L-N figure shown by the meter.

My gut feel is that summing the results for individual circuits (by resistances in parallel) would give a reasonable indication of the value to be expected from a full board test. Perhaps what we need is some good experimental data though.

- Andy.


Andy,
My calculation was to give an example, to the point that there is no relevance to trying to establish the sum of single circuit insulation values, I was not intending to produce a factual value. There will be multiple parallel paths in any installation; coupled with the fact that you can not measure the true insulation values of individual circuits. This would make any calculated value worthless.

-------------------------
http://www.niceic.biz
 04 June 2011 09:18 PM
User is offline View Users Profile Print this message


Avatar for sparkingchip.
sparkingchip

Posts: 5973
Joined: 18 January 2003

From my 1943 book to the 11th ed. regs. first carry out a insulation test between the two wires at the main switch, switches on, fuses in and lamps out.

Next as in my present avatar photo test between the two free ends of the wires at the main switch to the water pipe being used as the solid earth, the result should be around at least 4 Megaohms.

However: " The standard of insulation resistance that should be recorded as laid down by the IEE wiring rules, before a completed installation is permanently connected to the supply is not less in megaohms than 50 divided by the number of outlets (points and switches) from the fixed wiring, providing that it does not exceed 1 megaohm for the whole installation"

Maybe a balanced approach to allow the size of the installation to be taken into account, however the example in the book of a domestic house installation has 26 points and there is not a cooker point or socket installed in the kitchen, only a light!

Andy
Statistics

See Also:



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