Best way to deal with moisture in cable

[bhavik]

I have a 6mm cooker cable in the wall. Perhaps a choc bar connector has been used (don't know for sure). The cable is showing continuity after disconnecting each end and testing. This suggests moisture perhaps. (Lots of cats around so no mice or rats. )

Is it worth trying to clear the moisture by passing high voltage to heat the cable?
Any suggestions on equipment e.g. Robin

(Client had a similar problem 15 years ago when new kitchen was installed but don't know what was done.)

 

[bhavik]

I think that myself and the other members have concluded that you need to carry out some more comprehensive testing and stop being so vague in your answers.

Apologies, I should have posted after more thorough testing. Unfortunately, my tester (fluke 1652C) died and I hope to get a replacement soon.

 

[JBW175]

Apologies, I should have posted after more thorough testing. Unfortunately, my tester (fluke 1652C) died and I hope to get a replacement soon.

And I apologise for maybe coming across as being a bit harsh, but it’s a cooker circuit and in an ideal world, It should run from CU to isolator, then from isolator to cooker connection plate.

The very least you need to do is split it down at these points and then IR the cable. This will give you a good indication of which leg the fault is on.

Also, with regards to it being a cooker circuit, have you pulled the cooker out and checked the condition of the connection plate?

It could well be that after 15 years of water, soup, and other detritus falling down the back of the cooker, it’s got inside the plate and the fault lies there.

 

[westward10]

I doubt moisture would introduce a low enough resistance to show physical continuity.

 

[littlespark]

Neon on the cooker switch?

Random thoughts, haven’t read the rest of the replies..

 

[James]

Neon on the cooker switch?

Random thoughts, haven’t read the rest of the replies..

Op said cable disconnected both ends.

 

[bhavik]

And I apologise for maybe coming across as being a bit harsh, but it’s a cooker circuit and in an ideal world, It should run from CU to isolator, then from isolator to cooker connection plate.

The very least you need to do is split it down at these points and then IR the cable. This will give you a good indication of which leg the fault is on.

Also, with regards to it being a cooker circuit, have you pulled the cooker out and checked the condition of the connection plate?

It could well be that after 15 years of water, soup, and other detritus falling down the back of the cooker, it’s got inside the plate and the fault lies there.

Off at cooker switch. from cooker switch goes into isolator. Isolator connects to oven and microwave - no cooker. problem on cable coming in to cooker switch. cable going out from cooker switch to isolator and onwards to oven and microwave is ok.

 

[Lucien Nunes]

Original symptoms were RCD tripping. Haven't done a lot of testing yet. Continuity between live and neutral.

Continuity L-N won't trip the RCD. It will make a loud bang, spit out some molten metal, leave a burn mark and trip the MCB.

Anyway, let's suppose there's 'moisture' causing leakage L-E. We need to be clear what that 'moisture' is. A bit of condensation on the surface of a terminal block will not trip an RCD, it needs to be properly wet. What is the highest insulation resistance that can possibly trip an RCD? A typical RCD trips around 22mA, let's suppose there's 4mA of leakage already on it and it only needs 18mA from the insulation breakdown. That would put the insulation at 230 / 0.018 = 12.8kΩ. Considering that most circuits achieve 10,000 times better insulation or more, gives an idea of why that's spectacularly low.

0.018 * 230 = 4W of heat dissipated at the fault. That is enough heat to melt insulation and cause rapid corrosion. Then there is the subject of surface tracking. When significant leakage takes place through water over a period of time, metal ions are dissolved from the conductors and can be deposited on the surface of the insulation. This is very much more serious on DC but it does still occur on AC. When the moisture evaporates, a surface track of degraded insulation strength is left behind. If the problem is at terminals they are likely to be badly corroded and high-resistance. The last 'damp' fitting I saw that had just occasionally started tripping an RCD was a double-socket with free water in the back box that ran down the wall when I drew a Stanley knife round it to break the paint seal. It was so badly corroded that the terminal screw heads had rusted away and the conductors were jet black with copper oxide for 3cm inside the insulation.

It's often OK to dry out moisture if it's fresh and / or the parts have not yet been subjected to mains voltage. But it is not safe simply to dry things out when they have likely been simultaneously subjected to moisture and voltage for an unknown period.

 

[bhavik]

Continuity L-N won't trip the RCD. It will make a loud bang, spit out some molten metal, leave a burn mark and trip the MCB.

Anyway, let's suppose there's 'moisture' causing leakage L-E. We need to be clear what that 'moisture' is. A bit of condensation on the surface of a terminal block will not trip an RCD, it needs to be properly wet. What is the highest insulation resistance that can possibly trip an RCD? A typical RCD trips around 22mA, let's suppose there's 4mA of leakage already on it and it only needs 18mA from the insulation breakdown. That would put the insulation at 230 / 0.018 = 12.8kΩ. Considering that most circuits achieve 10,000 times better insulation or more, gives an idea of why that's spectacularly low.

0.018 * 230 = 4W of heat dissipated at the fault. That is enough heat to melt insulation and cause rapid corrosion. Then there is the subject of surface tracking. When significant leakage takes place through water over a period of time, metal ions are dissolved from the conductors and can be deposited on the surface of the insulation. This is very much more serious on DC but it does still occur on AC. When the moisture evaporates, a surface track of degraded insulation strength is left behind. If the problem is at terminals they are likely to be badly corroded and high-resistance. The last 'damp' fitting I saw that had just occasionally started tripping an RCD was a double-socket with free water in the back box that ran down the wall when I drew a Stanley knife round it to break the paint seal. It was so badly corroded that the terminal screw heads had rusted away and the conductors were jet black with copper oxide for 3cm inside the insulation.

It's often OK to dry out moisture if it's fresh and / or the parts have not yet been subjected to mains voltage. But it is not safe simply to dry things out when they have likely been simultaneously subjected to moisture and voltage for an unknown period.

Well explained. Good basic maths. Will do more testing in a few days. From the sounds of it, the cable will likely need replacing.

 

[bhavik]

IR testing:
Live to Earth is only 0.01 mega ohms (10 KΩ) when testing at both 250V and 500V.

Live to Neutral and Neutral to Earth are both 200 mega ohms (250V), 500 mega ohms (500V).

 

[ipf]

Suggest you find the choc bar connector to see if it's melted.

 

[JBW175]

IR testing:
Live to Earth is only 0.01 mega ohms (10 KΩ) when testing at both 250V and 500V.

Live to Neutral and Neutral to Earth are both 200 mega ohms (250V), 500 mega ohms (500V).

Looks like it’s time to start lifting up floors and digging into walls to find the problem then.

At least you know what part of the circuit it’s on after splitting it down and IR testing it.

 

[bhavik]

Suggest you find the choc bar connector to see if it's melted.

Would like to do that. But may not be practical as part of the cable is in the wall. If the connector was found (for example in a suspended ceiling), what next?

 

[James]

Would like to do that. But may not be practical as part of the cable is in the wall. If the connector was found (for example in a suspended ceiling), what next?

Take a photo of it and post it in dodgy trade pics thread.

then cut back the cable to the undamaged part and using correctly rated maintenance free connections repair it.
test
certify
give happy customer the invoice

goto local pub for a pint or local cafe for cheesecake

 

[littlespark]

goto local pub for a pint or local cafe for cheesecake

Save time and find a pub that serves cheesecake. Or a cafe that sells beer…..

Or cheesecake flavoured beer.

I’m calling it!

 

[UNG]

Would like to do that. But may not be practical as part of the cable is in the wall. If the connector was found (for example in a suspended ceiling), what next?

You need a cable tracer, should lead you to where the fault is

 

[bhavik]

Take a photo of it and post it in dodgy trade pics thread.

then cut back the cable to the undamaged part and using correctly rated maintenance free connections repair it.
test
certify
give happy customer the invoice

goto local pub for a pint or local cafe for cheesecake

https://community.NoLinkingToThis/threads/extending-the-existing-6mm-cooker-cable.52161/

Yellow in line connectors:

4.0mm² - 6mm² Yellow In-line Butt Connectors

 

[Lucien Nunes]

I don't trust no-brand insulated crimps on 32A circuits. Plus, you'll need to join the 2.5mm² solid CPC and most crimps are not suitable for solid cores (some are, but they are specific types and you must use the matching tooling which can be £££)

Would suggest three Wago 773-173's which do 2.5 to 6.0. and are rated for 41A:

Wago 773-173 Push Wire Connector 2.5mm - 6.0mm 3 Way | Wago_Connectors (773-173)

 

[bhavik]

Continuity readings:
L E. 0.15 ohms
N E 0.88 ohms

 

[bhavik]

You need a cable tracer, should lead you to where the fault is

Any recommendations on cable tracer?
Most will trace the wire. I presume the signal would weaken around the fault. There are loads on market from cheap to the expensive fluke 2042

 

[Lucien Nunes]

Continuity readings:
L E. 0.15 ohms
N E 0.88 ohms

Really ohms?

This thread started off all about moisture but now we get onto the actual readings they are perhaps 10,000 times lower than moisture is likely to be responsible for. Unless it's raining mercury?

That's a proper short-circuit you have there, a nail through the cable or similar, and all three conductors involved too.

 

[James]

Really ohms?

This thread started off all about moisture but now we get onto the actual readings they are perhaps 10,000 times lower than moisture is likely to be responsible for. Unless it's raining mercury?

That's a proper short-circuit you have there, a nail through the cable or similar, and all three conductors involved too.

Raining mercury, sounds mad as a hatter to me!!

 

[westward10]

You must mean meg ohm?

 

[UNG]

Any recommendations on cable tracer?
Most will trace the wire. I presume the signal would weaken around the fault. There are loads on market from cheap to the expensive fluke 2042

I've not spent too much time looking at the current crop of cable tracers if I was in the market for one now it would probably be a Fluke or Ideal Industries unit

My goto for this type of fault would be my 3M cable tracer which was bought in the mid 80's and long since discontinued back then it cost around £500 and has well earned it's keep over the years, it operates on with voltages from 9 - 600v AC/DC so can be used on live and dead cables with a battery supply

 

[bhavik]

Really ohms?

This thread started off all about moisture but now we get onto the actual readings they are perhaps 10,000 times lower than moisture is likely to be responsible for. Unless it's raining mercury?

That's a proper short-circuit you have there, a nail through the cable or similar, and all three conductors involved too.

Really ohms?

This thread started off all about moisture but now we get onto the actual readings they are perhaps 10,000 times lower than moisture is likely to be responsible for. Unless it's raining mercury?

That's a proper short-circuit you have there, a nail through the cable or similar, and all three conductors involved too.

I tested at fuse box end after connecting the live and earth at the other end. Same for neutral and earth.

Normal is about 0.72 ohms according to above.

 

[westward10]

Are you saying you have done R1+R2 and R1+N ?

 

[UNG]

So you have done a continuity test which is not that relevant to finding the fault you mention in the OP, you really need to be doing an IR check between the conductors

 

[bhavik]

Are you saying you have done R1+R2 and R1+N ?

No. Just connected live and earth at switch end and then tested for continuity. Then similarly for Neutral, I connected neutral and earth at switch end before testing continuity at fuse box end.

I posted IR test results earlier.

I tested as shown in link below

 

[bhavik]

So you have done a continuity test which is not that relevant to finding the fault you mention in the OP, you really need to be doing an IR check between the conductors

I was thinking: could continuity tests help to indicate whether fault is nearer the fusebox end or the other end.

 

[bhavik]

No. Just connected live and earth at switch end and then tested for continuity. Then similarly for Neutral, I connected neutral and earth at switch end before testing continuity at fuse box end.

I posted IR test results earlier.

I tested as shown in link below

With both ends disconnected, the continuity reading is maximum ohms for all three tests: LE, NE, LN

 

[Lucien Nunes]

Continuity readings:
L E. 0.15 ohms
N E 0.88 ohms

IR testing:
Live to Earth is only 0.01 mega ohms (10 KΩ) when testing at both 250V and 500V.
Live to Neutral and Neutral to Earth are both 200 mega ohms (250V), 500 mega ohms (500V).

With both ends disconnected, the continuity reading is maximum ohms for all three tests: LE, NE, LN

The continuity readings should be similar. If they are both correct (which they quite probably aren't) then the high one appears too high, because for 6mm² T&E cable R1 +R2 should be around 0.01Ω/m. 0.88Ω implies the cable is over 80 metres long which surely isn't correct. 0.15Ω looks entirely plausible, therefore the excess of 0.63Ω could indicate a high-resistance joint in the neutral that will heat dangerously under load (at 20A that's 252W, the heat of ten soldering irons)

The insulation resistance L-E at the fault seems to fall in the middle of the two ranges on the tester, off the bottom of the scale for IR (reading 0.01MΩ or lower (you can't tell how low when the display is showing the lowest possible number other than zero) but off the top of the scale for the continuity test. This is a limitation of most MFTs but we don't really need to know the actual resistance as it's definitely no good as it stands.

So what we can infer from the readings if they are all correct is that the insulation of the line is seriously low, possibly low enough to overheat dangerously, while the continuity of the neutral is seriously high, possibly high enough to overheat dangerously. The readings don't enable the location of the damage to be narrowed down.