Different continuity R1 and RN readings on ring

[Motty]

Changing a CCU and reconnecting existing wiring (twin and earth) = did a N - CPC IR test for earth leaks prior to dismantling the old (BS3036) unit for obvious new RCD trips; all ok.

However, on ring circuit the continuity end-to-end readings of the R1 and Rn are 0.3 ohms different; 0.41 on live, 0.72 on neutral. did the obvious and took every socket off to check the N terminals, all seemed ok. all fuses on spur outlets pulled out, at sockets the R1+R2/4 (calculated 0.22, checked out / recorded as a 0.33 at a spur) appears normal.

Other than an inaccessible pinched / stretched / kinked section of cable, I'm unsure of what could be causing this. MFT behaving itself fine on the second ring with identical LL and NN readings. I appreciate the more critical is the R1+R2 but not happy about the different L-L, and N-N continuity readings. Interconnection seems eliminated by r1+r2 readings appearing normal. Enlightenment, or memory jog by kick up the backside, would be very welcome!

cheers

 

[telectrix]

may sound stupid, but are you sure you measured rN and not r2 to get 0.72? other than that, it's got to be a loose/dirty termination on the N. could be a JB under the floor somewhere.

 

[Lucien Nunes]

I would test using a wander lead, leg by leg until a point is found that R1 and Rn get out of step, and look for strangeness nearby. If the difference of 0.3Ω is all concentrated at one point it will stand out clearly against the low resistance of each leg of line. Sounds like duff connection as Tel says, you would have trouble damaging a cable so as to make such a difference, without severing it completely!

 

[Motty]

ta for replies, the customer had laid a new carpet and wooden flooring so taking these up to investigate further - the JB possibility - may be a thory issue ... I'm thinking the best way to alleviate is if I split the ring at the faulty link I can then make two 20A radials? there'll be enough spare ways in the board to do this.

 

[SirKit Breaker]

As Tel says, check that you haven't measured r2 instead of rn, the figure of 0.72 is about right for a r2 measurement if your r1 is 0.41, for T&E cable. Failing that, have good clean of the cable ends, and make sure you have a good tight connection with the test leads, and also make sure you nulled the leads between r1 and rn tests. This has got to be an easy fix, i doubt very much that you will find anything by pulling the house to bits.

Cheers………….Howard

Cheers…………….Howard

 

[Tridentsystems]

Quote of the day ^ look for strangeness nearby ^ love it

 

[Motty]

Honest guv the cable that measured .072 was the neutral, CPC came in at 0.66. Have had a word with the customer and they're happy for the ring to be split into two radials if need be to compensate, as they are solely bedroom sockets only. I shall be looking for strangeness galore! but will be stripping the whole lot down and re-terminating to be sure. nice 1 for the replys

 

[Guest 004]

So you have readings that comply with the regs, and yet you want to pull the house apart to find a dirty joint?

 

[Lucien Nunes]

A joint registering 0.3Ω, or a cable with one live conductor nearly twice the resistance of the other, is faulty. Does a faulty cable comply with the regs?

 

[Guest 004]

A joint registering 0.3Ω, or a cable with one live conductor nearly twice the resistance of the other, is faulty. Does a faulty cable comply with the regs?

The IR was clear according to the OP.

 

[Guest 004]

I would suggest NOT calculating the R1R2, and measure at each socket with your cross connected cables. First with phase earth, then with phase neutral.

 

[Motty]

I would suggest NOT calculating the R1R2, and measure at each socket with your cross connected cables. First with phase earth, then with phase neutral.

Calculated and measured, end-to-end and cross connected as per usual procedure. on phase-CPC got 0.33 at a known spur. everything else was around 0.22-.26 mark, all fascias were taken off looking for a loose neutral screw. maybe i'm just being pedantic but I would rather the LL and NN readings near enough identical. as its bedroom sockets I've already agreed with the customer if the faulty leg can't be solved without major disruption I will split ring to two 16 or 20A radials to be safe, maybe over cautious for some but I'm one of lifes' worriers!

- - - Updated - - -

 

[Guest 004]

Did you do R1RN? At each socket with cross connections.

This should highlight if you do actually have a problem.

As suggested earlier your readings are ok, we do not live in a perfect world. Otherwise we could calculate everything.

If you still cannot get the reading down to similar as the phase conductors, just make a note of it on the minor works certificate.

 

[Guest 004]

Point to add as well is that before you do any remedial works, the circuits should be tested before hand to highlight any problems, as you will become liable, which unfortunately you are finding out now.

 

[Lucien Nunes]

The IR was clear according to the OP.

The IR might be wonderful but I never mentioned it.

A joint with 0.3Ω resistance, if that is the cause of the odd reading in the neutral, is capable of getting quite hot. E.g. passing 10A it would dissipate 30W and get as hot as a soldering iron. At 32A it would be giving out ten times as much heat. As it's a ring, it would cause more or less incorrect sharing of the total load between the two sides instead, depending on where it's located.

The world is not perfect as you say but it is better to find out the imperfections with the MFT than by tracing the burning smell.

 

[Guest 004]

Changing a CCU and reconnecting existing wiring (twin and earth) = did a N - CPC IR test for earth leaks prior to dismantling the old (BS3036) unit.

cheers

The resistance would need to be a lot higher buddy.

 

[Guest 004]

The IR might be wonderful but I never mentioned it.

A joint with 0.3Ω resistance, if that is the cause of the odd reading in the neutral, is capable of getting quite hot. E.g. passing 10A it would dissipate 30W and get as hot as a soldering iron. At 32A it would be giving out ten times as much heat. As it's a ring, it would cause more or less incorrect sharing of the total load between the two sides instead, depending on where it's located.

The world is not perfect as you say but it is better to find out the imperfections with the MFT than by tracing the burning smell.

That is just bad math

 

[Lucien Nunes]

Eh?

P=I²R
=10 x 10 x 0.3
=30W

or
=32 x 32 x 0.3
=307W

Just to make sure we're on the same page, I'm working in base 10.

 

[Guest 004]

Are you seriously suggesting that a 2.5mm cable cannot handle 307W?

 

[Silly Sausage]

No he's not.
We're talking about a poor joint here, so in few mm[SUP]3[/SUP] of metal, dissipating 100s of Watts, things are going to get HOT!

 

[Guest 004]

Poor joint or poor accessory

 

[Lucien Nunes]

I see we're thinking along similar lines now. Another theory was that one side of the cable had been 'stretched'. For a bit of mental arithmetic practice, try the following:
A single-core cable of length L1 with resistance 0.41Ω is stretched so that its cross-section reduces uniformly until the resistance is increased to 0.72Ω. What is its new length L2, assuming the resistivity of the copper remains constant*?

* it does change somewhat, when you work metals.

 

[Guest 004]

Ok we will have fun with that, but the install is twin and earth.

 

[Lucien Nunes]

the install is twin and earth.

My point exactly. You would have to do bizarre things to the cable to make that difference. When the OP has a chance to do the wander lead test or R1+RN/4, we can discover whether it's a rogue reading, a bad connection, or whatever. In the meantime, L2/L1= ?

 

[Guest 004]

I see we're thinking along similar lines now. Another theory was that one side of the cable had been 'stretched'. For a bit of mental arithmetic practice, try the following:
A single-core cable of length L1 with resistance 0.41Ω is stretched so that its cross-section reduces uniformly until the resistance is increased to 0.72Ω. What is its new length L2, assuming the resistivity of the copper remains constant*?

* it does change somewhat, when you work metals.

2.5mm is 7.41 Milli ohms per metre
So length 1 is 55 metres
And when stretched is 97 metres

Leaving resistivity out of the equation

if you stretch a cable the cross sectional area of that cable will reduce thereby increasing the resistance/metre to an unkonwn value

hence it will not have to be as long!!!

 

[Lucien Nunes]

it will not have to be as long

True

increasing the resistance/metre to an unknown value

False

Hint: You can leave CSA and actual length out of the equation too, to find L2/L1. Volume of copper is constant. Sub an expression for CSA into the expression for R. Finding numerical answer will require taking a square root.

It was just to show the extreme stretchification needed to account for an extra 0.31Ω

 

[Motty]

Really appreciate all the replys, and the different perspectives. Sorry didnt get back til now internet been off for 2 wks thanks to a change in phone provider!

I explored the legs of each socket by stripping all fascias off and plotting the ring layout. However, found that the ring must go to a JB under the floor, because the boiler switched fused isolator and 2 sockets in the 2nd bedroom were all spurs from this inaccessible box.

Went back over what I had checked and after L-L(0.41 ohms) N-N (0.71ohms) CPC-CPC(0.66ohms) testing, the R1 + R2 readings appeared normal (0.27-9) on the 4 sockets or first "half" of the ring, and 0.36 ohms highest at the beroom 2 spurs. Incidentally the R1+RN readings at each outlet varied between 0.21 to 0.30 when cross-connected. Incidentally IR readings came in at >999Mohm L-N, 857Mohm L - CPC, >999Mohm N - CPC.

Customer insists I cant take up the flooring upstairs. For now, all sockets re-terminated, and put a 16A MCB to lower the risk of overload on the circuit. Planning to put on the EIC for the CCU change that IMHO this ring should be rewired and re-configured, and next inspection 5 yrs rather than 10. As rightly stated, I am liable fot the safety of this circuit now i've changed the CCU, and can't seem to stop worrying about it ... A sharp learning curve out in the real world!

 

[Badged01]

What is its new length L2, assuming the resistivity of the copper remains constant*?

Missed this gem of a post originally but couldn't resist an early morning mathematical challenge .... I get the new length ~ 73M.

Correct or are my maths failing me again?