Ideas for finding fault area in underground swa?

[HappyHippyDad]

Afternoon everyone..

During some testing I have found an IR reading of 0.01MΩ between N-E. This is on a length of SWA buried underground from the main consumer unit in the house to a garage consumer unit. The poor readings are not on the outgoing circuits in the garage.

There must be a join somewhere as it is T&E leaving the main CU and then SWA entering the garage CU. I cannot find this join.

The length of run is very approx 40m (as I don't know the course it takes underground). It is below many paving stones, and various other obstacles. It would cost a lot and cause major disruption to run a new cable so I am considering how to find the fault area. It's a difficult choice as it could takes days of fault finding and perhaps cause even more disruption than a planned route for a new cable.

One thought is to use an ezicat 100 cable finder to at least work out the route of the cable and then dig in certain areas (that are easier to get to). Then perhaps cut the cable and test both ends to see which I can keep in situ and which needs further investigation. I have never used an ezicat 100 before, are they reliable? Would this be a situation where it is the right tool for the job?

EziCAT 100 - Trent Hire and Sales

It detects alternating (AC) electromagnetic signals radiated due to currents flowing in conductive buried services. The operator is given the location and direction of the underground service when the unit receives reradiated radio frequencies through internal aerials, which detect the strength...

www.trenthireandsales.co.uk

Thanks all.

 

[happysteve]

Is the reading low enough to detect on continuity, and if so, is it low enough that it's a dead short? If so, this will give you an idea of distance from each end of the fault, but not necessarily where this is. Has to be a dead short to be remotely accurate.

A cable finder/locator is the right sort of tool, definitely. Presumably it comes with a genny (generator/transmitter) to generate the signal that the receiver will detect.
I've not used these sorts of tools personally, yet. I've just bought a Fluke 2042 detector+transmitter - a much smaller-scale set-up - for a job next week (cable buried in wall rather than underground)

 

[James]

As above, you can estimate by measuring the resistance of the fault from both ends but it is not always spot on.

 

[davesparks]

One thought is to use an ezicat 100 cable finder to at least work out the route of the cable and then dig in certain areas (that are easier to get to). Then perhaps cut the cable and test both ends to see which I can keep in situ and which needs further investigation.

This approach is a gamble, you might end up adding the cost of hiring a cable locator and your time to the cost of replacing the cable in its entirety.

If the cable has been damaged underground and left like that for some time then there is a high chance that water has got into it. If water gets under the sheath of SWA it will spread along the cable and rot through the armour for quite a long way.

 

[littlespark]

T&e when it leaves the board, but no visible joint?

More than likely the t&e is all the way and stops short of the garage. If it was swa all the way, the joint would be local to the board…
Just thoughts on jobs I’ve been on where DIY Dave has just spurred off a kitchen socket, no isolation and run out to a shed in t&e buried in a hosepipe.

A couple of exploratory digs where cable leaves house, or enters garage, in hopefully easy to dig ground.
Double check under floor at board for joint or is there a socket nearby board that was maybe an original radial circuit, and they’ve come from there with the armour.

 

[OnlQQker]

Could save you thousands and the peeps a lot of anxiety.

Moling Equipment Hire | Plant Hire | Speedy Hire

Impact Moling Specialists - Mole Group UK

Impact Moling is a trenchless method of laying pipes or cables underground, where a pneumatically driven Impact Mole forces a path underground.

moleutilities.co.uk

 

[Avo Mk8]

You may need a combination of the detector you have mentioned, to find the route of the cable, and a TDR cable fault locator to give you a measurement of the distance along the cable to the join and/or the fault. You can hire TDR equipment, though not cheap.
I'm not familiar with the TDR equipment that's available these days I'm afraid.

davsparks makes a good point about the possibility of corrosion damage.

 

[OnlQQker]

Ps, just for reference , If you see any watermains people, (not going to mention names) but the guys digging up the watermains.

It's worth chatting to them because they'll mole it for you, a lot cheaper and quicker than you think.

 

[littlespark]

Speak to anybody that’s dug up the ground for water mains…. They probably nicked the cable with a digger bucket

 

[timhoward]

If you fancy some fun the old school way would be to use a wheatstone bridge and either a Varley loop test for a very long cable (if total resistance or 2 cores is over an ohm) , or Murray loop test if under an ohm.
Everything needed is neatly packed into a wind-up Megger BR3 or BR4 tester. Any serious collector of junk who's an electrician should have a BR4....!

I've had success with this method once in terms of being guided to the right couple of meters and then spotting the obviously newer fence post. However then I hit the exact issue that @davesparks mentioned - after chasing corrosion in both directions the cable then went underneath a tennis court that they didn't want to disturb. Game over!

 

[HappyHippyDad]

Well, I have a bit to think about then!

My gut feeling is it would just be better to run a new cable, but it is such a difficult run to make neat and it is a very well kept garden/grounds. It would be good to at least use some of the existing buried SWA, the majority of which is likely to be absolutely fine.

A dilemma.

Thanks for all the replies.

 

[timhoward]

In terms of finding the cable route - I've had very good success with this relatively inexpensive tool:

Pardon our interruption...

I had a silly situation once where I had to prove to the DNO that a bungalow wasn't fed from a substation on the private estate. Using one of those I traced the supply cable underground, under a boundary wall and to the middle of a road using one of those. I could even pick up the signal at the nearest neighbours cut-out, proving they were on the same phase.
You'd need one core connected to live, and the tool connected to that core and the earth stake it comes with.

 

[littlespark]

Have you tried two bits of bent fence wire held loosely in your closed hands?

 

[littlespark]

Have you tried two bits of bent fence wire held loosely in your closed hands?

You might laugh..... but on the caravan site, that was more reliable at finding cables than our Ezicat... (something else i wish i'd "squirrelled away" when we sold up)

Considering the Ezicat is designed for cable avoidance and not cable finding.... And our cables, drains and water pipes tend to utilise the same trenches.

 

[Lucien Nunes]

A loop / bridge test will probably find the fault. TDR will probably find the joint and/or the fault if they are in different places. You might recall my Heath Robinson TDR setup that I wrote a thread about, using my laptop and an external VGA monitor when I didn't have an actual TDR instrument available, which was capable of locating a fault to within a metre or two. I'd still use that method now, or at least try wielding some technology at it first before digging randomly along the route.

You mention the IR reading as '0.01MΩ' which covers a multitude of sins as it's effectively off the scale of the IR tester. If the resolution of that range is 0.01MΩ and it's allowed +/- 1 count, that could be anything between 0Ω and 15,000Ω. A continuity reading that gives more figures would help characterise the fault.