Unexplained RCD Trips

[confuzed]

Hi All,

Hoping someone can give me some idea of where to investigate:-

I'm getting occasional (ie about once per fortnight) RCD trips, which happen at events on non-RCD circuits. I'll explain....

I have a split load CU, on a TNS system. Main and supplementary bonding is to 16th edition. Non RCD side of CU has cooker, immersion heater (turned off), smoke alarms, garage, central heating, two lighting circuits.

RCD side has upstairs ring, downstairs ring, kitchen ring, utility room radial on it.

The last 2 RCD trips have happened on the following circumstances:-

- Turning the oven off
- Turning off a light upstairs, bulb blew, RCD tripped.

Both of the circuits are on the non-rcd side. And i'm 100% certain the RCD tripped at the exact moment the two events happened.

Any pointers for investigation much appreciated.

~Conf

I'm certain there is no crossed neutral, all circuits have IR > 200Mohm. I'm at a loss. Any ideas?

 

[EddieCurrent]

Hello Confuzed,

I find this post very interesting and wondered if you had had any joy elsewhere in getting any advice ????.

(I thought that it was going to be a incorrectly connected neutrals, untill I got to the end of you post, where you explain that you are certain it is not).

Now, I am confused too.

 

[THRIPSTER]

Perhaps the switching/fusing transients induced currents in adjacent circuits (the RCD side circuits) sufficient to cause imbalance in the phase/neutral ccts and cross the switching threshold of the RCD?

Same sort of thing as when you hear switching 'plops' on radio's/HiFi's.


Regards

 

[confuzed]

Thanks for the comments. I'm no closer to figuring the problem out.

Even though I've checked that the neutrals are in the right places, I still can't see how a crossed neutral would cause the fault I described.

Surely a crossed neutral would cause the RCD to trip when a device on the incorrectly connected circuit was switched on? (provided that device pulled more than 30mA)(which a cooker or a light would definately do).

Does anyone else have any thoughts at all? Pointers for what to look out for?

 

[confuzed]

Perhaps the switching/fusing transients induced currents in adjacent circuits (the RCD side circuits) sufficient to cause imbalance in the phase/neutral ccts and cross the switching threshold of the RCD?

Same sort of thing as when you hear switching 'plops' on radio's/HiFi's.


Regards

I can see, in theory, how that could work. But surely it'd be a relatively common problem?

 

[EddieCurrent]

Hello,
Glad to see the post finally moving...

Is it possible that the lighting circuit could have a 'borrowed' neutral somewhere, and the lamp blowing caused the trip in that case ??

 

[confuzed]

anyone else got any ideas?

 

[davelx]

Do you have a lot of audio, home cinema, PCs etc on the rings, or some form of surge suppressors etc?

Most modern electronic equipment has filter capacitors between phase, neutral and earth that cause a small amount of current to flow in the CPC. If you have enough of this type of kit, the earth current may approach the RCD tripping level making it more sensitive to transients.

It might be worth unplugging everything possible when not in use (the input filters tend to be before the mains switches) to see if that makes any difference.

 

[JohnD]

Something else that may contribute:

If you have a generally high level of earth leakage (say, 25mA which is not quite enough to trip the RCD) from the various appliances on the RCD protected circuits (especially watery ones like washing machines, kettles, and outside lights and sockets with rain water in them) then it takes very little to push the breaker over the edge, and a tiny transient might do it.

It is difficult to track down these intermittent trips without testgear and some practice.

 

[confuzed]

Thanks for the comments.

I do have 2 PC's and a laptop. I know that one of the regular PC's was on when the cooker-turn-off trip happened, but I can't be sure if they were at other times.

I've previously tryed measuring current in the earth conductor that runs from the CU earth terminal to the MET, but read 0.0 (via a clamp meter). But that's inconclusive - clamp meters are not terribly accurate, and there could be leakage current down a different path to earth.

The PSU in that PC is of the cheapo-made-in-china variety. I think i shall be naughty and a bit wreckless and test it's leakage current with an ammeter in series with it's earth.

What specialist test equipment do you refer to JohnD ?

 

[securespark]

You need a clamp meter that measures mA.

You also really need to get the RCD tested to see that is tripping when it should in the correct timeframe, and not when it shouldn't.

 

[confuzed]

my clamp meter does measure mA. I'll have a continued play tonight.

cheers

 

[confuzed]

correction, clamp meter does not measure mA.

However, multimeter does. All my PC equipment together leaks 1.73 mA.

But that is just one circuit, can't really get the leakage on the others unless i put my multimeter in series with the MET. Which probably isn't the safest idea in the world ever.

-------------------------------------------------------

Ok, so just put the meter inbetween the main earth block and the TN-S earth, meter in peak hold mode, maximum current was 6.62 mA. I went round the house and turned on and off most devices.

This to me doesn't seem an unreasonable amount of leakage current for an entire installation. So I'm no closer to figuring out what's wrong.