RCD Trips when lights turned off !

Andrew Mawson

unread,

Feb 4, 2013, 2:12:43 PM2/4/13

to

New installation. Four 'uplighters' each equipped with a G24-D2 bulb:

http://www.saving-light-bulbs.co.uk/energy-saving-2pin/18w-push-fit-cfl-g24d-2-pl-c#.URAFaB002So

Simple lighting circuit two way switched, sourced from the same breaker as
other light circuits on that floor. Within two weeks the RCD for that bit of
the installation has tripped five times when the lights are turned off. RCD
was new, has been tested for trip point and time. Circuit (and whole
installation) has been tested using up to date, in cal, certified 17th Ed
equipment. Needless to say I cannot get it to trip on demand. Seems (but
only an impression) to happen when lights have been on for a while).

Presumably there must be an imbalance on switch off sometimes causing this,
but why not every time? Any one experienced anything like this, or has
suggestions what may be happening? Back EMF surge from the ballasts or
something? - I'm puzzled and so is my sparks.

AWEM

Dave Liquorice

unread,

Feb 4, 2013, 3:24:54 PM2/4/13

to

On Mon, 4 Feb 2013 19:12:43 -0000, Andrew Mawson wrote:

> Simple lighting circuit two way switched, sourced from the same breaker
> as other light circuits on that floor.

New installation shouldn't be a borrowed neutral and I'd expect that to
reliably trip not just "random" and after two weeks. The latter maybe due
to other kit sharing the same MCB appearing and being a bit leaky though.
So there is less headroom in the RCD for extra leakage.

Faulty uplighter? Connect only one and work through them one by one.

--
Cheers
Dave.

Owain

unread,

Feb 4, 2013, 3:33:05 PM2/4/13

to

On Feb 4, 7:12 pm, "Andrew Mawson" wrote:
> Seems (but
> only an impression) to happen when lights have been on for a while).

Temperature-related distortion of some component? Neutral-earth short?
Borrowed neutral?

Owain

Andrew Mawson

unread,

Feb 4, 2013, 4:21:58 PM2/4/13

to

"Owain" wrote in message
news:0fa23f2f-6977-4416...@e18g2000vbv.googlegroups.com...

I think borrowed neutral is almost certainly ruled out as it only happens
maybe once in a hundred switching offs. (As Dave noted) Quite possibly a
warming up of the chokes I suppose, but why only failing at the instant of
turn off? Google did reveal a chap in Kent who had a very similar problem
with some G24-D2 garden lights. He solved his problem by putting a snubber
capacitor / resistor network (0.1uF in series with 47 ohms) across switched
live and neutral so I may do that and see what occurs - but this shouldn't
be necessary for an off the shelf B&Q light! I have another four of these
lights in anther room with no issues!

AWEM

ARW

unread,

Feb 4, 2013, 5:42:20 PM2/4/13

to

The first thing I do with a problem like that (assuming you have not got
immediate access again to the test gear) is to stick on a 500W load to see
what happens.

--
Adam

Brian Gaff

unread,

Feb 5, 2013, 3:44:30 AM2/5/13

to

Sounds like back emf to me, though have to say, it is not a common thing to
occur. I suppose if you judged it just as the max voltage was on the
ballast, then the back surge would be greatest.

Brian

--
From the Sofa of Brian Gaff Reply address is active
"Andrew Mawson" <andrew@please_remove_me.mawson.org.uk> wrote in message
news:ML2dneoAkMC2lY3M...@bt.com...

Andrew Mawson

unread,

Feb 5, 2013, 7:45:42 AM2/5/13

to

"ARW" wrote in message news:kepdgd$753$1...@dont-email.me...

As in "tune for minimum smoke" I suppose!

AWEM

meow...@care2.com

unread,

Feb 5, 2013, 7:14:26 PM2/5/13

to

When the switch is opened, it can arc momentarily. The resulting hf causes much increased current through the usual capacitance to earth.

Options: fit a snubber to the switch, or reduce the total load on the rcd.

NT