The Engineer's Notebook Blog

### The Engineer's Notebook

The Engineer's Notebook is a shared blog for entries that don't fit into a specific CR4 blog. Topics may range from grammar to physics and could be research or or an individual's thoughts - like you'd jot down in a well-used notebook.

 Previous in Blog: The Epidemic That Wasn't (Part 2) Next in Blog: Solving an Electrical Fault from 10,000 Miles Away (Part 2)

# Solving an Electrical Fault from 10,000 Miles Away (Part 1)

Posted November 09, 2009 12:01 AM by PWSlack

Editor's Note: PWSlack had an "annoying electrical fault" in the United Kingdom. Masu lives in Australia, but provided his fellow CR4 guru with "encouragement and suggestions" from 10,000 miles away. Every day, CR4ers such as PWSlack and Masu collaborate and communicate through our community's open forums and private mail system. What follows, courtesy of PWSlack and in his own words, is evidence of one such effort that is too good not to share. - Moose

"Why does the oven keep tripping the earth trip? It's so annoying. Can you sort it out?"

Good questions.

The bothy is fed with 230V 50Hz electricity from overhead cables, and TT is the earth system in use. There is an earth rod at the 11kV-230V distribution transformer in a field some distance away, which is connected to Neutral at the pole. 230V Live and Neutral are fed to a brick connection cabinet in a neighbour's field. Inside the cabinet are the kWh meter and a 100A supply fuse together with an isolation switch. The outgoing cable is a steel wire armoured one, with two of the 4 cores connected to Live and 2 to Neutral. The cable disappears underground before surfacing inside the bothy, where both Live and Neutral pass through a 100A/100mA RCD before being connected to an 8-way distribution board with 100A double-pole isolating switch before going onto the individual circuit breakers.

All Earth conductos witin the building are connected to an Earth bar inside the distribution board. The Earth bar is connected to two earth electrodes at different places outside the building.

The oven circuit is fed via a 32A/30mA RCBO, which trips properly when the "test" button is pressed.

What was curious was why the 100mA RCD took the incoming supply out, whereas had there been a fault in the oven, the 30mA RCBO would have gone instead. So there had to be an earth fault, somewhere downstream of the distribution board, which leaked more than 100mA to earth when the oven was approaching full throb and less than 100mA when only a couple of rings were on.

CR4's Masu was helful in providing some guidance and encouragement. The thoughts went round and round and round...

Then it clicked.

The words of the professional Electrician that carried out the testing and certification of the recent bothy extension work electrics in compliance with Part P of the UK's Building Regulations a couple of years ago came ringing back: "You've got a good Earth!" Well, with an Earth-to-Neutral fault elsewhere within the bothy, that expression would make sense, and correlated well with the oven tripping the supply out when taking a high current from the supply. Masu was right. There was a sinking feeling. Tools, old clothes and overalls <sigh>….

Chapter 1: Investigation

Having isolated it and fired up the multimeter, the oven checked out OK at its incoming terminals, with >20 megohms between the protective Earth conductor and both the Live and the Neutral conductors, no matter how many cooking rings, grill elements and oven elements were switched on at any moment. The same story was told at the distribution board with the oven reconnected and the meter connected to the oven circuit cable disconnected at the board. No problem there.

The distribution board is in a cupboard in the hall. With the incoming mains isolated, there isn't much light in there, so temporary lighting needed to be brought into play. A head-mounted LED torch with wind-up charging to its internal batteries proved to be a godsend; both hands could be kept free while there was good lighting wherever one's head is turned at the same time.

There was still a fault.

"Have you fixed it yet?" Mutter, mutter.

With the supply still isolated, the next step was to meter for ohms between the Neutral terminal and all the Earth conductors running through the circuit cables within the bothy, which were connected to an Earth bar within the board. 0.3 ohms or pretty much a hard short circuit. Hmmmm. The next step was to disconnect the circuit Earth conductors from the Earth bar, one at a time, and see if the meter reading changed. Nothing happened until one of the two earth conductors that were part of circuit no.6, the 6A lighting circuit, was disconnected, at which point the meter went >20 megohms. Bingo! Somewhere on one of the two cables feeding the lighting circuits, Neutral and Earth were touching!

The errant cable feeds about half the bothy lighting fixtures, some of which had metal casings including all of the light switches. So leaving the earth wire disconnected at the distribution board was not a good idea, as in the event of a Live to Earth fault developing on the circuit, all of the metal fittings would have become live until the fault current exceeded 100mA, which is enough to kill a human. So the logical step was to hook up a temporary feed for the errant cable, with the Earth conductor disconnected for now, from a BS1363 plug with 3A fuse plugged into a wall socket near the distribution board via a plug-in 30mA Power Breaker (usual disclaimer) auxiliary RCD. At least there was now an combination of a level of safety, a lighting circuit that sort-of worked (OK, so the plug's 3A fuse blew once when an ageing filament lamp downstream of it went through its death throes) and an oven that could now be operated at full throb without the supply going off.

Editor's Note: Part 2 of this series will run tomorrow.

Interested in this topic? By joining CR4 you can "subscribe" to
Guru

Join Date: Aug 2006
Posts: 3408
#1

### Re: Solving an Electrical Fault from 10,000 Miles Away (Part 1)

11/09/2009 10:55 PM

I think it was a very nice engineering advice by Masu.

__________________
Prof. (Dr.) Shyam, Managing Director for Sensors Technology Private Limited. Gwalior, MP474001, India.
Power-User

Join Date: Aug 2009
Location: Gulf Region, IIEE_CRCSA
Posts: 113
#2

### Re: Solving an Electrical Fault from 10,000 Miles Away (Part 1)

11/10/2009 3:46 AM

What is the exact supply voltage(KV) of 10,000miles short transmission lines.

Give us the electrical data

1. Line %impedance, %reactance etc. or shall we assume values.

The we can calculate the electrical fault.

__________________
If there is a will... there is a way... No pain / No gain.
Guru

Join Date: Feb 2009
Posts: 837
#3

### Re: Solving an Electrical Fault from 10,000 Miles Away (Part 1)

11/10/2009 10:05 AM

PWslack ,as I have observed in many forums is a great guru, contributor and educator. By discussing, sharing views and brain storming ,the exercise will give new insights and exposure to all members. I am not an electrical engineer anyway, but the volunteer ship and meeting challenging problems and solving ,makes good sense on the part of both the experts [PWslack & Masu]. My hearty best wishes to the initiatives. POLISHING KNOWLEDGE BY INDULGENCE AND APPLICATION- the most interesting job for creative technicians.

__________________
Nature is so graceful and naked. Human possession is ridiculous.
Guru

Join Date: Apr 2007
Posts: 3531
#4

### Re: Solving an Electrical Fault from 10,000 Miles Away (Part 1)

11/10/2009 1:49 PM

Sorry to have missed the original thread...
If PWSlack's bothy ("basic shelter") is equipped like this, I'd really like to see his house.

But the description of the wiring sounds like a tracing nightmare - at least compared with our superficially similar fault in a (new-to-us) cottage. The fault became apparent after we found that the old 100-mA RCD had failed and replaced it with a 30-mA version. Fortunately, our ground spike and Faraday cage connections were reasonably close to the distribution board. So the offending circuit could be found using found by local loading rather than by isolation every circuit separately. The problem was in one of the 30-amp rings, so we could identify the offending one of the 18 sections by successive halving. 4 steps (good - it could have taken 5) took us to the offending link. We simply removed live and neutral (leaving the ground looped) and downgraded the MCB for that ring from 32 to 20-amps (never yet exceeded - although we do have space on the board for another 20-amp MCB).