RCD Tripping - Outdoor Lighting Circuit - Only on connect / disconnect

Orson100

Hi all,

I have been researching RCDs and how they operate and activate etc, quite a bit over the past while, but I cannot find a post or discussion that is relevant to my particular issue.

In a nutshell: I have an outdoor lighting circuit with 4 x fluorescent light receptacles, 3 x 11w and 1 x 15W or thereabouts, and 2 x motion detectors in the mix also, all SWA cabling mostly buried, all underground electrical junctions made off and sealed in IP-rated boxes, and all of the lighting receptacles similarly IP-rated for outdoor use. The circuit is terminated in a standard 13A plug which I connect to one of the standard sockets in the house. My issue is that around 50% of the time, when said plug is either connected or disconnected, the RCD trips in my MCB.

Please note that once the RCD is flipped back up, the lighting circuit will remain energised, and will work no problem for days on end.

I've left it energised overnight on several occasions, including during heavy rain, morning dew etc. and no issues. Based on my limited knowledge, the fact that it runs okay in this challenging environment would seem to indicate that this is not an issue of moisture ingress to any of the junctions or a problem with insulation resistance in general, is this a reasonable assumption?

To re-iterate, issue only ever occurs when I am plugging / unplugging the circuit so I am probably doing something suspect at this point and would appreciate some guidance on best practise or input on what might be the cause?

TIA
Orson.

Risteard81

A number of points:

Firstly I'm not sure what type of fluorescent luminaires you would have externally. Secondly you mention IP rated boxes however do the luminaires for example have drain holes to enable moisture to actually escape after changes in temperature etc. cause condensation?

Thirdly an SWA cannot ever be fitted into an IS 401 (BS 1363) plug top. These are designed ONLY for a flex with a cross-sectional area not exceeding 1.5mm^2. The SWA could not have been fitted correctly and will also not have earthed armour without the gland which is essential for the protection it is to afford.

However, the RCD problem could also be related to the switch contacts in the socket outlet. Really the only answer is to get a Registered Electrical Contractor. The whole setup sounds extremely poor, non-compliant and possibly dangerous.

Hard-wiring it would also have been far more sensible than having a plug and socket arrangement.

Orson100

@risteard81, thanks for taking the time to reply. I'll get an REC to have a look at the setup.

Re your questions about fluorescent luminaires and outdoors, all I can say is that I bought the external lighting fixtures from a reputable lighting store, paid a good bit, and the receptacles take fluorescent bulbs, so I can only assume that they are fit for purpose.

Thanks
Orson.

2011

Orson100 wrote: »

I have been researching RCDs and how they operate and activate etc, quite a bit over the past while, but I cannot find a post or discussion that is relevant to my particular issue.

RCDs installed in houses are generally set to trip with a current imbalance of 30mA.

Without getting too technical:
In a properly functioning circuit the magnitude of the current flowing in the phase (live) is equal to that returning in the neutral. If there is a difference in the return current of > 30mA the RCD will operate (trip).

Assuming that your RCD is functioning correctly you have a current imbalance issue. So where is the "missing" current flowing?
Answer: Generally to earth.
Why? Because the insulation is not what as good as it should be.
Why is the fault intermittent? Because sometimes the current flowing to earth is greater than other times.
Why? It could be because moisture ingress is greater at sometimes than other times. Also the RCD may be looking at the current a number of circuits and the combined current to earth varies depending on the loads connected at a given time.
Why dose this happen more often when you connect / energise the lights? This
could be because lights that contain a choke have a certain amount of leakage current even when operating properly and this combined with the leakage from other circuits pushes it over the threshold.
Therefore if these lights were on a dedicated RCD the problem may go away.
What other circuits are on the RCD? Frequently multiple socket circuits in a domestic installation are protected by a single RCD.

underground electrical junctions made off and sealed in IP-rated boxes, and all of the lighting receptacles similarly IP-rated for outdoor use

Are the junction boxes designed to be buried? What IP rating, IP44, IP66 ?
Most likely this is where the problem is. Are they glanded into the boxes? If so what glands? Is the armour connection to earth maintained?

I would only use resin filled "submarine" type junctions on SWA for underground. Personally I wouldn't trust much else.

The circuit is terminated in a standard 13A plug which I connect to one of the standard sockets in the house.

This would be considered bad practice for permanently connected circuits.

My issue is that around 50% of the time, when said plug is either connected or disconnected, the RCD trips in my MCB.

Does the MCB trip?

I've left it energised overnight on several occasions, including during heavy rain, morning dew etc. and no issues.

Perhaps it takes several hours for the water to filter down or the other circuits contributing to the current imbalance are not drawing a load.

Based on my limited knowledge, the fact that it runs okay in this challenging environment would seem to indicate that this is not an issue of moisture ingress to any of the junctions or a problem with insulation resistance in general, is this a reasonable assumption?

No, see above.

all I can say is that I bought the external lighting fixtures from a reputable lighting store, paid a good bit,

Means nothing unless you know what you are buying. Do the lights fittings have a metal or plastic enclosure?

If I was the electrican dealing with this (I don't work as an electrican) I would start with:

● Carry out an insulation resistance test on all circuits that are fed by the RCD. Investigate anomalies accordingly.
● Test the RCD. Perhaps it is over sensitive.
● Connect the lights properly (via light switch).
● Supply the lights from a dedicated RCD or better still an RCBO (RCD and MCB combined in a single unit).
● Ensure that underground electrical connections and made off properly.
● Ensure that all installation work is compliant with the regulations.
● Use quality materials.

In my opinion it is time to get a professional in. It is best to get a REC that you know / trust or is recommend because standards vary wildly.

Orson100

@2011, wow, thanks for the in-depth reply, much appreciated.

To confirm to begin, yes I am going to get a registered contractor to have a look, I am out of my depth.

I don't know the answer to your questions about how the boxes were prepared, I did not bury them but I will found out.

I had my terminology wrong: an MCB (miniature circuit breaker) has never tripped, always the 30mA RCD.

I am still of the opinion that I don't have an issue with moisture ingress though; I can identify no temporal link (whether immediate or delayed) between incidents of rainy wet weather and the incidences where the RCD trips when I connect or disconnect the circuit. As I said, the circuit will run for days on end during wet and wild without incident, so surely if there is a leak to earth that is being aggravated by moisture ingress events in the circuit, then at some point in time the RCD will trip *while* the circuit is energised, and not solely at the point when it is either connected or disconnected?

I think you are more on track with your reference to a cumulative leak towards the 30mA threshold with the lighting circuit pushing it over. I have many sockets with many four-gangs with many, many appliances, handheld computers, PCs, TVs, routers, power supplies etc, many of which are plugged in all the time. All of the house sockets are all on the one RCD.

One thing that puzzles me is that this anomaly also occurs when I *dis*connect the lighting so if anything, would this event not contribute to a reduction in the leak to earth (if it exists) thereby reducing the risk of the RCD tripping? (Even As I write this, I know there is probably some electrical explanation for this too.)

I think another, if not a dedicated, RCD will be the way to go, but I will be advised by a professional from now on.

Thanks again for all of the information.

Rgds
Orson.

2011

Mod note: Please read the forum Charter before posting.

In particular this bit:

Can I promote or advertise a business or organization here?
In short, no. Advertising for free is not permitted on boards. Frequent "plugging" of a commercial enterprise, organization or product is not permitted.
However, as it is appreciated that many professionals offer advice here FOC, if you wish to post a discreet link in your signature, that is not a problem. Please note that a sticky is now in place on the forum, where contractors who are also Boards.ie members may be listed. Drop a PM to the mods if you would like to be included on this list. No discussion of work carried out at cash prices allowed.

See link:
http://www.boards.ie/vbulletin/showthread.php?t=2055192317

air

OP, you have gotten good advice.
It is likely that the inductive load of the fluorescent ballasts is causing the tripping which would explain why you only have issues when plugging in and out the circuit.
As already stated it should be on a fixed permanent circuit anyway.