Tripping Mains circuit - how to isolate?

Hire a certificated mains appliance mains leakage tester for a day and test everything on the wall outlets one by one - I think the equipment is called a PAT tester, its an acronym for something but I don't know what!

The only way the MCB could be tripping before the individual circuit CB is that the MCB is detecting more than overcurrent. If it has arc detection then there is a real and possibly dangerous condition with the existing house wiring or the culprit may be an old style wornout bimetal thermostat switching the mains directly- this would be the boiler and the refrigeration stuff. Plan on building a multichannel fast acquisition unit to capture those circuits first- current transformers, signal conditioned, and smart "pre" triggering is a must- obviously triggered on MCB trip. You might also contact the engineering department of the CB maker.

I'd change the MCB for another, as a first step

Is it an ELCB, or what ever they call it these days, it might be worth measuring the leakage current, it might just be on the limit. I got an old elektor scan somewhere with a cct for such a device

martin

Hi Terry...

Know absolutely nothing about the UK systems out here in the colonies (Canada); but just a thought for whatever it may be worth?

IF what you're tripping is ground fault protected, then surely the first place I'd be looking is at the water/high humidity devices - you specifically mentioned a pond pump.

FWIW, I have an individual ground fault interruper outlet on the side of my house - for the car, etc. Even a few drops of snow or rain on it is likely to cause it to "pop"

Take care.

Ken

I've had to replace a small submersible pump in a fish pond because of leakage current tripping the breaker.

--
Service to my country? Been there, Done that, and I've got my DD214 to
prove it.
Member of DAV #85.

Michael A. Terrell
Central Florida

There are several sensitivities of GFI, and those associated with pools and especially hot tubs can be extreme.

But the general solution is to disconnect half the loads and see if the problem persists. Then disconnect one quarter of the problem loads.

I had to do this once with a city fire alarm system. Which would trip only on very windy days. The problem took months to solve.

Finally found a short mid block and twenty feet in the air where the steel cable on a power drop had sawed its way through the alarm insulation.

--
Many thanks,

Don Lancaster                          voice phone: (928)428-4073
Synergetics   3860 West First Street   Box 809 Thatcher, AZ 85552
rss: http://www.tinaja.com/whtnu.xml   email: don@tinaja.com

Please visit my GURU's LAIR web site at http://www.tinaja.com

What's the sensitivity of the breaker to earth leakage current ?

If you have lots of smps powered products on your ring you might be surprised at the accumulated earth current from the Y caps in the EMI filters. Maybe it's just on the point of tripping all the time ?

Graham

First of all, I have to admit that I'm no expert on UK wiring systems and codes. With that aside, here's my input:

[snip]

Its doubtfull that a DMM will find anything other than a hard fault. If its intermittent, it could be something nonlinear, i.e. system voltage causes something to break down that will look OK at low voltage.

If I understand ring mains correctly, this is a distribution branck circuit that has two sources of power. Unlike radial circuits (used here in the USA), fault current will flow in from both sources and will be present on each segment of the circuit. The device you build will have to detect power magnitude and direction.

I don't know if this is safe or complies with code, but you could try to break the ring and place some overcurrent indication on each source. I'm not sure how a ring circuit breaker is configured (it would probably break both sources for an overcurrent on one side), but fuses would make the problem simpler.

With all that said, don't follow my advice unless you know it meets local codes!

--
Paul Hovnanian     mailto:Paul@Hovnanian.com
------------------------------------------------------------------
>> Insert witty message here

A ring main has a single source (the fuse for that circuit) its simply both ends of the ring are connected to the fuse, the ring main can have any number of radial spurs.

Sure all the above is probable but heres another. If you have a spur going to an outside shed or similar then lightening can suplement the earth current an trip the earth return. Had to bypass the trip on my water pump circuit for this reason.

Thanks for all the helpful replies so far. I'm leaning towards the view that these are what in further reading late last night I've seen called 'nuisance trips'. IOW maybe something (apart from me) showing signs of its age. Such as my 15 year old cooker (electric oven, gas hobs), 14 year old washer, 18 year old gas boiler, etc.

Graham's suggestion about SMPS-powered devices is another possibility. Presumably the obvious contributors are my PC and Epson Inkjet printer? (Not sure about other candidates like: flat-bed scanner, NiCd/NiMh battery charger, SW Radio, Realistic 2006 Radio Scanner, etc?) But if that was a serious issue wouldn't I have heard a lot more about it? FWIW, Graham, this Crabtree consumer unit has a prominent label '8A/30mA', so I assume the sensitivity is 30 mA?

Ken: I have a separate breaker unit on the house socket that takes power outside to the pond pump. That has never tripped, so I think I can exclude that as culprit.

The installing electrician (I 'phoned him last night) said one possibility was something to do with fluorescent lights. He's mid-European, English is not his first language, so he's rather hard to understand sometimes. (In the same conversation, when confirming I could do some leakage tests myself with my DMM, he referred to his own testing instrument as '250 Watt', when I assume he meant '250 M Ohm'.) So I may not have grasped that issue about fluorescent lights properly. But I think I can rule it out anyway based on no one being in the house on the last occasion.

The 'RCD Protected' circuits are apparently:

- Hot Water Immersion Unit (always switched off at wall)

- Upstairs mains sockets

- Downstairs mains sockets

- 'Office' (treated separately because of all the stuff here)

- Cooker

Fred was the only one who addressed the second part of my post, about some sort of automatic detection device. Any other thoughts on that please? On reflection, if the residual current triggering this thing is the cumulative effect from those 5 circuits above, then I'd need to interpose the detector in each one somehow, yes? But, given that I'm obviously not the first consumer with intermittent tripping, there must be commercial tools around?

The electrician has promised to call in today. I'll report back on progress. Meanwhile, any further advice would be much appreciated please.

--
Terry Pinnell
Hobbyist, West Sussex, UK

How about popping down to B & Q or wherever, and picking up one of those in-line ECLB sockets used for isolating outdoor portable equipment like lawnmowers ? These often have a higher sensitivity than 'proper' MCBs, so if you went round the house putting it ahead of each piece of plugged-in equipment in turn, you might well find that it trips before the MCB on that circuit, back at the consumer unit.

Just as a matter of interest, we used to have a similar problem at one of my wife's cafes, that was in premises shared with another business. We finally tracked it down to that other business's electric kettle. A pinhole of corrosion in the outer jacket of the element, is a known common problem for causing this. Electric toasters are another favourite.

Arfa

While SMPS's can trip an RCD, I'd expect it to do it a lot more often than every few weeks/months, and at turn-on. The cooker is way more likely to be the culprit, though you haven't mentioned if any/all of the trips happened with the washer going (the water being the obvious reason). Another poster's comment about looking for auxiliary causes (rain, wind, people kicking cats, whatever) is a good one - see if there's another common cause going on.

You could maybe break the circuit and run from two breakers, but that may not be possible if the current rating of the cabling then is insufficient. Possibly start with separating out the office and running it off its own RCD.

I don't know of any commercial equipment that will help you with this because of the intermittent nature. Regarding your sparkie's comment about a

250W tester, that may be the maximum power rating of the equipment under test that his tester can handle. I'd be surprised if it was fluoro lights, if only because I assume your wiring is similar to Antipodean wiring and the lighting is on a separate circuit to the power outlets (which are the ones with the RCD). If the lights are also protected maybe a ballast is breaking down.....? Oh, and you want to use a Megger to test the insulation, though at a couple of hundred pounds (RS price) you probably would want a real reason for buying one. :)

Cheers.

Ken

Hi Paul...

For the benefit of the OP, may I add that if he does go the route of fusing or GFI'ing one of those spurs, that it still won't necessarily tell him anything.

The one that opens is going to be the one that is either/or more sensitive to a ground fault, or failing that the one that's faster :)

Take care.

Ken

[snip]

That modifies the problem slightly. A fault on one of the radials can be detected by fusing or putting some sort of overcurrent indication on each. But for a problem on the ring itself, both sides of the loop will carry part of the overcurrent. So either the ring must be split or overcurrent detection placed in the middle of the ring must also sense power direction.

--
Paul Hovnanian     mailto:Paul@Hovnanian.com
------------------------------------------------------------------
When I was in high school, I remember boys and girls slept
together all the time. We called it algebra class. -- Jay Leno

I did suspect the (rather ancient) washing machine when it happened a few weeks ago. But as I said in my OP, *nothing* was being actively used on the most recent occasion.

During his visit yesterday I realised he had been saying 'Volts', not 'Watts'. ("Votz".) His tester has settings for 250V and 500V. Which of course is why it's much more suitable than my DMM; although that sports a handy 200 MOhm range, its 9V presumably provides an inadequate test of real conditions.

Turns out that was a tripping cause he'd experienced a couple of times, but only when light was actually being switched on/off. Inapplicable here.

Anyway, happy to report that cause now found. The overall 'rating' when he first applied the tester to the entire house was borderline, which of course was consistent with such an intermittent fault. Eventually we isolated that 'iffy' circuit to external garden wiring which I'd installed a decade ago. Switching to the Ohm meter, we got erratic readings, typically about 20 M. Interestingly, I'd tentatively excluded that garden circuit. It was one of a couple (my shed/workshop being another) that was supplied via another 30 mA RCD unit. But of course, as both Ken Weitzel and my electrician pointed out, the consumer unit breaker that did trip was probably already sensing a small leakage from everything *else* in the house. And/or it may be faster acting.

After his departure I started tracking down the actual fault. I aim to complete that today, but I'm pretty sure it will prove to be damp getting into one or more of my connections. (There are a couple of external mains sockets, linked by cable running from house. One of them takes 240V to a relay box at bottom of garden, which switches lights on at night.) Despite generous use of 'gaffer tape' (which will be a pain to remove), I expect rain has somehow found a way in!

Thanks to all here for the help and advice, which was much appreciated.

--
Terry Pinnell
Hobbyist, West Sussex, UK

than

be

happened

poster's

cats,

insufficient.

about a

the

ones

breaking

though

Glad you have (or soon will!) find the problem. Rather than gaffer tape, get hold of some heatshrink tubing, the type with the 'glue' in it - much better weatherproofing.

Cheers.

Ken

Ken Taylor scrobe on the papyrus:

. .

Self-amalgamating tape is great. We used miles of it in permanent cable installations for outside broadcasts. Never had a problem.

RS stock number 494-433.

--
John B

I like heatshrink tubing for neat joints (although I've never used any containing glue). But for joints of several wires of different sizes and condition, some with outers, possibly also using a screw-type connector, it's not versatile enough.

--
Terry Pinnell
Hobbyist, West Sussex, UK