One light circuit is tripping the system. what to look for?

[saveasteading]

One light circuit is tripping the system.  What to look for?

 

I have narrowed down the area of the problem through gradually switching on more fuses. and then trying one light switch at a time.

Tried taking the light bulb out on the grounds of wishful thinking.

It is a single ceiling bulb, with a simple one-way switch.

 

What should I do next?

 

My hunch is that mice have gnawed a cable in the attic, because we can hear them up there.

In a way that would be good news as it may be simple to resolve too.

 

On the other hand, there are may reels worth of exposed cabling up there, and this hasn't happened before. 

 

Or is the pendulum cable connection the first thing to check?

Do wall switches malfunction?

.....or what else?

 

[BotusBuild]

If you think a critter has gnawed a cable, then you'll need a tester to be able to check insulation --> time for an electrician.

If you suspect a switch, check connection tightness of cables on the switches and/or swap each switch in turn with a known good switch (tip: mark up the good switch so you know which is which after you get distracted 😉)

While you're at it, check cable tightness on all the light housings as well. As above, by swapping out with known good units.

[saveasteading]

If my logic is correct, the switch is simply a local circuit coming down from a rose ( or similar) above the light cable. If the switch is throwing a fuse, then the fault is in the cable over that short distance. Hopefully an exposed bit of it.

Thus I'm thinking it must be an issue at the connection above the light or the cable from there to the switch. (why would this happen after 20 years? and I haven't been near it to trip on it.)

Or a critter , again they don't seem to have been interested in the plentiful cabling over all these years.

 

Maybe it would be simpler to disconnect at the light connection and see if the tripping stops. could circuit test from there too. I am always nervous with live cables so would rather only test an isolated circuit.

 

I don't know if it's feasible to test within the switch box.

 

It looks like an unpleasant crawl through the attic is beckoning.

[BotusBuild]

After 20 years cable can deteriorate (unlikely but possible), and cable connectors can loosen. Don't ask me how but it happens.

Whenever I get a problem like this I start with the simple things - tighten the connections, then swap out units, then call a qualified electrician to run circuit tests (that way you're not touching live shit!)

Often one of the two solves the issue.

Edited Wednesday at 18:18 by BotusBuild

[SteamyTea]

Can you, safely, rewire at the ceiling rose to bypass the switch, then see if it trips?

 

When electrical testing it is not unusual to disconnect circuits and then bridge (twist the wires together) the circuit so that both sides (+&-) can be tested together, from one place.

[joe90]

What kind of fuse is it? More modern stuff tests fir all sorts if conditions unlike the old thin melting wire jobby in my day 🥸

[SteamyTea]

8 minutes ago, joe90 said:

unlike the old thin melting wire jobby

I still have them, and not blown one in the 2 decades I have been in the house.

They are probably in the original 1987 bits of fuse wire.

[saveasteading]

1 hour ago, joe90 said:

What kind of fuse is it?

I don't understand fuses. This one can trip at one of 2 places. I might send photos tomorrow. 

 

Before we had these 'improved' boxes installed ( I can't work on your house until they are changed") we had the old wire ones and they seemed to battle through the minor issues. 

[MikeSharp01]

If you can work out the switch connections in the ceiling rose you can disconnect there and insert a test switch but only play if you have an earth leakage breaker in the consumer unit - if not leave it alone. (Just getting ready for the online harms bill don,'t you know!)

[SteamyTea]

3 minutes ago, saveasteading said:

I don't understand fuses

They can be a bit confusing because of the terminology.

There are 'circuit breakers' that are just modern fuses i.e. they disconnect when the current goes to high for the circuit.

Then there are RCDs, residual current devices, theses are for safety and disconnect the supply, very quickly, if they sense an imbalance between the current going into a load, and the current coming out.

Then there are RCBOs, residual current circuit breakers, which combine the two above.

Now we also have AFCIs, arc fault circuit interrupters, these sense any sparking in a circuit i.e. a live wire that has been nibbled by a mouse, touch a neutral wire that has also been nibbled.

 

To makes it more confusing, the above can have different disconnect speeds, usually in milliseconds.  This it to allow for high inrush loads (usually an inductive load like a highly loaded motor starting up).

 

Old fashioned wire fuses are a bit more sophisticated than just a thin bit of wire.  They are designed, chemically, to allow a bit of over current for a short period of time.  But even if the load is below the designed capacity i.e. 5 amps, they will still blow given enough time.

The reason I am still on the original fuses is because my loads are very small.  Take my lighting circuit, this has gone from a total of 5 amps down to 0.1 amps, if I put all my lights on at once.  The most likely fuses to go are my cooker/oven and DHW circuit.  The DHW amperage has not changed i.e a nominal 3 kW.  I do limit the time it runs though, so rather than possibly on for 7 hours, it is on for 1.5 hours a day.

So it is not really a case that mechanical fuses are more robust/forgiving, really a case that if designed correctly and used sensibly, they can last decades.

 

 

[SteamyTea]

11 minutes ago, MikeSharp01 said:

Just getting ready for the online harms bill

That was being spoken about on the radio this evening.

@Pocster is going to be in big trouble.

[Mattg4321]

It does sound like it’s probably rodent damage. Especially if you can hear them! Should be straightforward for an electrician to locate with an insulation resistance test. Hopefully if access to cables in loft is ok it might be a simple fix too. 
 

The only thing you could try is to turn off the lighting circuit and get up there an expose the wiring and inspect for damage. Turn it off first though! I’ve seen it plenty of times where rodents have stripped cables bare!

 

You probably need to ignore most of the well meaning advice on here as half of it is nonsense in this instance. Getting a good electrician to sort is going to be the easiest and quickest fix here

[joe90]

51 minutes ago, Mattg4321 said:

You probably need to ignore most of the well meaning advice on here as half of it is nonsense in this instance.

What we’ll meaning advice do you consider nonsense ?

[G and J]

5 hours ago, BotusBuild said:

and cable connectors can loosen. Don't ask me how but it happens.

I don’t know but I’ve always assumed it’s due to temperature cycling - my loft temperature varies wildly over the year.  Perhaps there’s an ideal torque that a screw connector should be tightened to to avoid it but I’m damned if I know what it is.  
 

Anyway I’ve got the hump now so I’m withholding my well meaning advice nonsense (until I see what sort of device is tripping out). 

[saveasteading]

40 minutes ago, G and J said:

until I see what sort of device is tripping out). 

Photos tomorrow.

[Mattg4321]

Apologies if any offence caused. Probably a poor choice of words. 
 

Point does stand though, although I didn’t really want to get into an argument by being specific but seems as though I’ll probably have to. Again no offence intended. 
 

A switch itself is very unlikely, if not impossible to cause nuisance tripping. Will be a waste of effort swapping them. Only if there was a fault between the switch and an earthed back box could it cause tripping. 
 

Loose connections won’t cause tripping. More likely flickering lights etc or even a fire. Going round randomly tightening connections will be a waste of time - unless you have AFDD’s fitted, but this is unlikely as they’re not in widespread use here yet. Although it’s very debatable whether they actually work on lighting circuits as the current is too low. Note that AFCI is an American term for the same thing. 
 

Would be handy to know if it’s an MCB or an RCCB or an RCBO where the problem is. Unlikely to be an earth leakage breaker, these are mostly obsolete and came before RCD’s (RCCB/RCBO’s), some of which are current operated (basically an RCD) and some of which are voltage operated (on the earth). These are not really safe for use. 
 

PVC insulation won’t deteriorate in 20 years unless subjected to very high loads or extreme temperatures etc. If in a domestic property it’s unknown how long it will last in good conditions as the earliest examples from the 60’s are still in good condition. 
 

Fuses/fuse wire very seldom blow through age in my experience, although it does happen. Very normal to change an old fuse board from 40+ years ago that appears to have most of the original fuse wire intact. Usually the lighting circuits that go when incandescent bulbs/lamps blew. Or the odd ring final circuit where they’d had a dodgy appliance perhaps. Cooker and immersion circuits the least likely to go in my experience. Also, I’m not sure it’s a good thing if you’ve got fuses that don’t disconnect in the event of a fault and just battle through!!
 

 

[saveasteading]

Attached photos of the apparatus. 

The old black box on the far left isn't an issue. Next along trips sometimes. The next along is a different subcircuit.

Picture 2 is inside the second box for completeness..not relevant though.

This power flies through the attic to the box in pic 3.

It can ping on the orange switch to the left. At the adjacent black 'fuses' I found which circuit was the problem and then which switch, which is currently taped over and out of use....hence the fuse turned on again.

 

[elite]

You need to keep splitting the circuit until you narrow down the cable with the fault. This will be a lot easier with a tester that can do insulation resistance testing. You can pick up used testers on ebay, but probably cheaper / easier to get a spark in

[saveasteading]

1 hour ago, elite said:

the cable with the fault.

Done that. It is a single light switch.

I haven't ventured into the attic yet as there are children around for a few days.

[elite]

2 hours ago, saveasteading said:

Done that. It is a single light switch.

I haven't ventured into the attic yet as there are children around for a few days.

Interesting, unusal for a switch to cause such a fault, but I hadn't realised it was an attic switch which I guess is more "exposed"

[saveasteading]

@SteamyTea wondering if you could continue your instructive, with reference to the photos.

I don't understand why one pings rather than another,  esp as the fault under discussion can trigger either.

 

44 minutes ago, elite said:

attic switch

Switch in a room, up to the attic above where we would hope any damage is fairly obvious.

[SteamyTea]

27 minutes ago, saveasteading said:

wondering if you could continue your instructive, with reference to the photos

I would like to, but for a change, I am going out tonight, very rare occurrence for me that is.

A quick look at them and the seems to be MCBs (mini circuit breakers which are simple over current devices, or modern fuse wire), with RCDs (residual current devices) on the incoming supply.

@ProDave may be better placed to identify the components.

 

It is always possible that the MCB has gone faulty.

[G and J]

 

I believe the one ringed in red will trip if there is an imbalance in the current in the live vs the neutral, i.e. some juice is leaking out of the circuit, usually to earth. (In this case probably through a still gooey mouse corpse - but the good news is they dry and as they do their insulation level increases till it doesn’t trip out any more).

 

The one ringed in blue will trip if too much current is taken.  Say if the mouse body is shorting between the switched wire (i.e. a live wire) and a neutral.

 

You could wait and see if said body (or bodies) dries enough to save you having to firkle through the loft.  But popping your head up to put some good snap traps up to reduce or remove the population would be a really good idea.  I’d not recommend bait, it encourages them in (they think it’s a yummy food source) and then they die and stink.  If you are going to bait, do it outside and reduce the population nearby.  Our experience is that that is the most effective long term approach.

[andyscotland]

@saveasteading just to be sure I've understood:  the box marked DB3 that @G and J has marked up is at the origin of your supply, and the power is fed from there to the multi-way split load board in your final photo?

 

And the trip that is going is the one G&J has circled red, knocking out the whole system?

 

As @G and J says, that one is an RCD and will trip on a very small imbalance between the current going out on the phase (live) conductor and coming back on the neutral. It's not possible to see in the photo what rating that is, it will likely either be 30mA or 100mA.

 

The one you have marked/switched off in the downstream board is an MCB and will only fail on overload (likely 6A if it's a standard lighting circuit).

 

Therefore you have a fault that is causing a very small amount of current leakage. This means either a high-resistance short between phase and earth (e.g. a mouse body, or degraded insulation) or possibly a dead short between neutral and earth : the neutral is not quite at zero volts so connecting to earth will cause a very small current to flow.

 

Assuming you haven't recently made any changes to the system, and have tried removing/replacing the lamp itself then the most likely explanation is either recent cable damage, moisture ingress to a connection point, or failure of a connected electronic circuit (which sounds unlikely if you have isolated it to only happening when a specific light is switched on even with the lamp removed).

 

There could be a couple of reasons why it is tripping at the origin rather than at the RCD protecting the right hand half of your downstream board:

 

* Possibly, both RCDs are 30mA and the same type. You would generally try to avoid wiring RCDs in series like that because there is no guarantee which will trip first on a fault. Usually I would either try to run cable between boards in a way that means it doesn't need an upstream RCD (eg using steel wire armoured cable) or use a 100mA time delay RCD upstream to increase the chances that the downstream RCD trips first. However it is not always possible/allowable to do this and so you get the nuisance that sometimes a fault will trip the whole installation.

 

* It may also be that the lighting circuit itself is only producing a very low leakage but there is leakage elsewhere in the installation which means when it all adds up at the origin it is just enough to trip the RCD. It's common for electronics (computers, LED drivers, smart stuff) to produce a small level of background leakage due to the way electronic transformers work.

 

In fact it is even possible that the lighting circuit has always had some cable damage causing leakage but only recently have you connected more electronics elsewhere and so it has now become a problem.

 

Ultimately the only thing you can really do is get a proper insulation resistance tester and try first to isolate whether there's any part of the system with a lower-than-expected live / neutral resistance to earth. You will need to be very careful to disconnect absolutely everything electronic before testing to avoid damage.

[FuerteStu]

Assuming the circuit holds, but trips when switched on, it either implies the returning 'switched live' is coming in contact with cpc (earth), or the lamp fitting itself. 

 

If its recent, my first thought rather than pest related, would be water ingress. 

 

Moisture inside the switch, or fitting. 

 

There is also the possibility of the lamp itself causing the trip. 

 

A picture of the fitting would be handy to diagnose further