Worrying earth problem

[Davisonp]

Hi all,

I have been looking into a suspected earth leakage problem.
I have purchased a Metrel earth leakage clamp, and started by measuring what I expected would be the earth leakage on the consumer unit by clamping the main earth into the board. The clamp metre was fluctuating rapidly and reading between 100mA and 150mA, so that confused me as at least one of the RCD’s should have tripped with that much leakage.
Then I clamped the Gas and Water bonding. One read between 100 and 150mA, the other read between 200mA to 240mA.
Then I clamped the main distributor’s earth lead from the incoming supply to the MET, and this time I read between 1000mA and 1250mA.
I switched off the consumer unit main switch and set the clamp meter to read amps not milliamps, and even with the consumer unit turned off it still read 1.2 amps.
I tried my Megger clamp meter and that confirmed that with the consumer unit main switch turned off, there is about 1.2A flowing through the main earth cable.

Later I tried the Metrel meter at my house and it read a steady 5.5mA current through the main earth between the supply earth and the MET. I then turned off my home consumer unit and the reading dropped to 0.5mA, which I assume is background and nothing to worry about.
I also tried plugging my kettle into my PAT tester and clamped the Metrel meter over an exposed earth wire (in an adapted extension lead) and the PAT tester said the leakage was 0.55mA using a differential calculation, and the Metrel meter said it was 0.47mA, so it only differs by 0.08mA so it looks like the Metrel meter is working correctly and reasonably accurately, and there really is 1.2 amps coming from somewhere in my client's house.

I am not sure what to do now. Does this mean there is a problem with the distributor's earth? Or am I misinterpreting the results?
Is this situation as dangerous as I believe it is?

 

[Knobhead]

Disconnect the water and gas bonds. Then take you’re reading on the service head to MET. You should then get a true reading without stray spurious currents. I’ve seen up to 25A on an imported earth fault but the RCD’s were quite happy as it wasn’t a fault on the bits they were protecting. See #13 http://www.electriciansforums.net/electrical-forum-general-electrical-forum/21227-tingley-taps.html

 

[mjvor]

If this house is on a PME supply then there will be neutral currents going to earth through the earthing and bonding conductors which as Tony says won't affect the RCD as these currents do not come from the final circuits. They flow from the incoming neutral, through the main earth to the MET then through the bonding conductors to earth and then back to the supply transformer. If your own house is TNS then you don't get that effect.
Did you try clamping the L+N together? That should be steady and will give you the total residual current for the house.

 

[Knobhead]

Did you try clamping the L+N together? That should be steady and will give you the total residual current for the house.

Good point, didn’t think of that one.

Imported earth faults are a swine to sort out, but at a few mA the DNO won’t be too bothered, as in my case 25A it gets them flustered.

 

[Davisonp]

Hi,

Just an update, and a question. I gave the DNO contact details to the client and told him what to say and they will be there within 24 hours (1.2 Amps is still enough to be considered serious).

Would the clamping of the live and neutral cables at the same time given accurate results in this particular situation? I had tested both RCD’s in the consumer unit, and both appear to be working fine (including a ramp test showing 19mA trip current), yet both RCD’s appear to be tripping as if they have a hair trigger. Before discovering the external earth current I had assumed either faulty RCD’s, or earth leakage from equipment causing the RCD’s to trip frequently, but is it possible that the external current in the earth cable is causing this phenomenon with the RCD’s tripping?
If the external current coming in on the earth is messing with the RCD’s, then it must be doing it by messing up the live or neutral current through the RCD’s. Does this make sense or have I misunderstood something (I am still learning)?

Once they have solved the external earth current, we will see if that fixes the RCD trip problem.

 

[telectrix]

your RCDs will trip due to a difference in the currents flowing in L and N, so, as mjvor said, if you clamp L/N together, any reading on the clamp meter will be the difference.

 

[Davisonp]

Another update.
The DNO engineers turned up at 9.00Pm on Saturday. The engineer's notes on the job notes say there was “No Significant Current between the head and MET”, so I went round at 9.30pm the following evening and measured 1.2 Amps again. Then I went back today to try and find the leakage problem, and there it was again. 1.2 Amps, so I guess the DNO engineer was unlucky and arrived when the problem was not happening.

I tried the idea of clamping live and neutral and measuring the difference (leakage) and that did work. I was easily able to see that my fears that the earth current might be leaking into the neutral cables was unfounded as I turned all the MCB’s off, and although there was still the 1.2 amps on the earth cable, there was 0.00mA on the neutral cable.
Then I set about turning on and off all the MCB’s and noting the leakage. It turns out that there is a problem on the Kitchen sockets (8.4mA leakage) and the total leakage including the kitchen sockets is 19.1mA, and the RCD test says the RCD will trip at 19.0mA, so it is right on the edge of tripping all the time.
I am trying to identify if it is a faulty item of equipment or if I need to be IR testing the circuit and finding the problem.

 

[Knobhead]

Good on you! But get the DNO back, they have a problem, not major but still not right.

 

[1shortcircuit]

Well done you for keeping your head and narrowing the fault :thumbsup

Another thread that shows an earth leakage tester is most definitely a must for any electricians tool box, it's not just a neat gadget

 

[mjvor]

You've not understood what I said above. Current in the earthing conductor is normal if you are on a PME supply. It flows from the supplier's neutral through the earthing conductor and through the main bonds back to earth. It is not causing the RCDs to trip. The only current which does this is the leakage current in those circuits. Use the IR tester to identify where the problem is. Could be water in a JB somewhere or damage to a cable. Make sure everything is unplugged.

 

[1shortcircuit]

You've not understood what I said above. Current in the earthing conductor is normal if you are on a PME supply. It flows from the supplier's neutral through the earthing conductor and through the main bonds back to earth. It is not causing the RCDs to trip. The only current which does this is the leakage current in those circuits. Use the IR tester to identify where the problem is. Could be water in a JB somewhere or damage to a cable. Make sure everything is unplugged.

Did you read the entire post? :thumbsup

It turns out that there is a problem on the Kitchen sockets (8.4mA leakage) and the total leakage including the kitchen sockets is 19.1mA, and the RCD test says the RCD will trip at 19.0mA, so it is right on the edge of tripping all the time.
I am trying to identify if it is a faulty item of equipment or if I need to be IR testing the circuit and finding the problem.

 

[Davisonp]

You've not understood what I said above. Current in the earthing conductor is normal if you are on a PME supply. It flows from the supplier's neutral through the earthing conductor and through the main bonds back to earth. It is not causing the RCDs to trip. The only current which does this is the leakage current in those circuits. Use the IR tester to identify where the problem is. Could be water in a JB somewhere or damage to a cable. Make sure everything is unplugged.

I forgot to mention in the first post, this is a TN-S system. The MET is fed directly from a cable attached to a clamp round the incoming mains.

I am curious about the amount of current running through a PME system. Any idea what size currents can be expected in a normal installation?

Also, Does the current pose any danger (Eg. Is it possible for the person servicing the boiler to provide an alternate path to earth and get a shock)?

 

[Davisonp]

Another update.

Two phone calls from the DNO today. The first to tell me that the engineer is refusing to go back to the property, so I asked to escalate this and the job details were passed to the shift supervisor.

The Shift supervisor should go into politics. He was incredibly evasive. Every time I asked about the current coming in on the earth he would either try and blame the customer side of the installation, or start talking about something that was unrelated to the problem like how the engineer tested between live and neutral at the head and there is not any problem there.
He never did officially acknowledge that there was 1.2 Amps coming into the property on the earth they supply.
I kept bringing him back to the problem, and eventually he told me what the engineer WOULD have tested.
The supervisor says that the current leaking into the property is “not the responsibility of the supply company”, and that as long as the Ze reading is a low enough reading, that is where their responsibility ends. Does anyone know any different?
He then suggested I request a conversion to a TN-C-S system which will take 8 to 12 weeks before it is done.
Is this the best way forward? Or should I be sending a video of me with the consumer unit turned off and the 1.2Amp of current flowing into the installation, and then asking the DNO to confirm in writing that they either consider this safe, or that they are not responsible for correcting this issue?

What do I do next?

 

[polo1]

Does the property in which you are having the problem share any metallic services (water, gas etc) with an adjacent property; is the incomming electrical service cable looped through an adjacent property?

Regards.

 

[Davisonp]

Does the property in which you are having the problem share any metallic services (water, gas etc) with an adjacent property; is the incomming electrical service cable looped through an adjacent property?

Regards.

I Don’t know.

Is there any way of telling the answers by testing or investigation? The client does not know.

 

[WDMDL]

For example with a pair of semi-detached houses, it is common to see looping through.
It's very easy to spot... if your service head has two cables, one will be the incomer and the other a loop to next door.
If your service head has only one cable, a quick nosey next door is required to see if theirs has two...

Simon.

 

[Davisonp]

Final update.

I sent an email to the customer services with a photo of 1.2 Amps coming down the earth cable with two direct questions. Is this safe? And who is responsible for making it safe.


The next morning I got a phone call from the distributors top trouble shooter who wanted to investigate himself.

He had some difficulty in identifying where the current was coming from, and tried PME’ing the supply 50 meters from the house, but when that did not change anything he came back again the next day with several new ideas.

What they actually did was:

1. Pull the distributors fuse, then separate their earth cable, from the MET.
2. With a multimeter, they measured the Amps between their earth and the supply neutral. It measured 1mA.
3. They then did the same between the MET and the supply neutral. 850mA this time.
4. Now, they connected the supply earth to the supply neutral, thus giving a very good return path and measured the current with a clamp meter. If there is current coming down their earth, it will flow directly down the neutral. The current was measured with a clamp meter and it measured 0.0mA.
5. Lastly, they connected the MET to the supply neutral, and this time they found that there was a current of over 3 Amps (The neutral has a reduced impedance compared to the earth so the current would be greater).
   This proves the current is actually coming from outside, through the house system and down the supply earth to the transformer.

Basically, someone in the area has a bad earth connection and is sending current through the ground, and this house has such a good connection to the ground that the current is using the house system as a return path to the transformer, but the voltage drop means the 1.2 amps is at less than one volt.

The Trouble Shooter put another PME connection in at the junction to the main power cable in the road, and was thinking about turning the house into a proper PME system, but eventually he decided the added dangers of PME were not justified in this situation, so the 1.2 Amps measured on the supply earth was actually less than 1 volt and does not pose any danger.

NOTE: The Trouble Shooter is going to retire in June, and has been working on the distribution network for over 50 years. In all that time he has never come across a situation like this. It appears to be a one off.

 

[1shortcircuit]

Davisonp, Thanks for the update:thumbsup

 

[polo1]

Davisonp, Thanks for the update:thumbsup

Ditto! Always good to hear how unusual problems are "solved".