Yet another RCD nuisance tripping topic

[stunlawless]

Morning everyone,

Had a bit of trouble recently. I've got an old brick barn (a shippon in Cheshirespeak), which is weatherproof but unheated.

Recently had trouble with with RCD tripping with two specific bits of kit. Both are cat 2, and both will trip the RCD if you touch the switch to either switch it on or off. If it's running, then it runs fine !

I've worked around it by switching on/off at the SO.

Install is TT with a 100mA TD at origin, shippon is fed by a submain to a DB, and then to a CU with all ccts protected by 30mA RCD. It's this one that always goes.

10 ways used in CU, including overhead supplies to two other sheds and outside lights. In the shippon are three light ccts with about 20 fittings (florries & pendants), and two SO ccts with about 20 points. There's a lot of small "rooms" and dark corners in this shippon. It's probably got as many outlets as an average 3bed house, but the load is lower.

Tested the RCD; times OK, tripped at 23mA on a ramp test. A little low perhaps, but not reason alone to change.

This happened a few weeks ago. Came to use the kit again yesterday, and no problems. The only difference is that now we're in a dry spell, and when I had problems, it was a wet spell.

I'm thinking that either the humid conditions led to a relatively high earth leakage in the fixed wiring etc, and somehome touching the switch just tops it off, or (and this is a wild shot) something to do with static. One item was a belt sander (belt rubs against plate all the time), the other is a vacuum that I was using to clean up the sanding dust (dust + motor ?). Could I have discharged a static charge which would confuse an RCD ?

No other kit showed the same problems, and it never trips unexpectedly.

Any thoughts ?

Thanks David

 

[bernardgreen]

Could I have discharged a static charge which would confuse an RCD ?

Possibly but it would be more likely to happen in dry conditions when static can build up to higher energy levels.

If it was just touching the metal case of the switch then I would be looking at how well that case was connected to the ground rod. It seems like your touch was adding a parallel path to ground thus reducing the impedance for fault currents to ground thus increasing fault / leakage current to earth above the threshold for the RCD.

 

[stunlawless]

Hallo Bernard,

Both items cat 2, all plastic body & switch case. Vacuum has a plastic earth pin, sander has 2 core flex. No connection to earth rod at all (except via the ground beneath my feet !).

I've just had another thought - maybe I became statically charged, which discharged through the machines live or neutral ? I was wearing rubber soled boots !

Surely the current from a static discharge that I didn't feel on the end of finger can't be anywhere the mA for 10's of ms range ? uA for single ms surely ?

It only happened when I touched the switch, which of course is where the live wire (maybe neutral as well if it's a DP sw - sold in Europe as well ?)

 

[bernardgreen]

Both items cat 2, all plastic body & switch case. Vacuum has a plastic earth pin, sander has 2 core flex. No connection to earth rod at all (except via the ground beneath my feet !).

You can put a CAT2 ( double insulated ) appliance into a plastic bucket of water and the RCD will not trip. It cannot trip as there is no path for any leakage to ground if the bucket is non conductive or standing on an insulating surface. The water in the bucket will be somewhere around 115 volts to ground. Only when you introduce a path for current from the water to the earth will there be un-balance in the live and neutral and if this is large enough the RCD should trip.

Maybe the vacuum and sander were damp enough that the insulation was bridged providing a conductive path the the area around the switch. Touching the case then extended that path to earth and this provided a path for earth leakage.

I've just had another thought - maybe I became statically charged, which discharged through the machines live or neutral ? I was wearing rubber soled boots !

Possible but many rubber soled boots and some shoes are intentionally made slightly conductive to prevent build up of static on the person wearing them. The impedance is low enough to leak away the static charge but high enough that if the person grabs a live wire then the current through feet and shoes is not high enough to be dangerous or even noticable. ( depends on the type of shoe of course )

Surely the current from a static discharge that I didn't feel on the end of finger can't be anywhere the mA for 10's of ms range ? uA for single ms surely ?

It depends on a lot of things whether a pulse of static discharge could trip an RCD. In theory it could as there can be a lot of energy in a person charged to a couple of hundred volts. A few hundred milliamps for an extremely short time could be enough to saturate the sensor and throw the trip. But not long enough for the nerves in the finger to react to it.

 

[stunlawless]

Hallo Bernard,

Definitely food for though there. My boots are Dickies Riggers, and described as "antistatic", but no DIN or EN or even BS referred to. Maybe they don't allow me to build uo a static charge, but will allow a discharge from something charges above earth (well, concrete floor) potential.

I reckon that, somehow, touching the sw creates enough of a path to earth to trip the RCD. Exaggerated in humid conditions on both of the machines, and some earth leakage on the fixed install.

If I had the right ammeter, I'd put it on the MEC when it's raining to see how close to tripping the RCD.

But, in the absence of any other nuisance trips, I'm going to wear rubber gloves when the weather is humid.

Thanks for your help, David

 

[cozycats]

I've worked around it by switching on/off at the SO.

Install is TT with a 100mA TD at origin, shippon is fed by a submain to a DB, and then to a CU with all ccts protected by 30mA RCD. It's this one that always goes.

10 ways used in CU, including overhead supplies to two other sheds and outside lights. In the shippon are three light ccts with about 20 fittings (florries & pendants), and two SO ccts with about 20 points. There's a lot of small "rooms" and dark corners in this shippon. It's probably got as many outlets as an average 3bed house, but the load is lower.

Are you using the RCD as a 'main switch' to switch all this?
If so therein lies the problem, if you really want one point of switching use the (I presume) mcb's in the DB individually rather than the RCD, hopefully that should be OK

 

[stunlawless]

Cozycats, not sure I get your point.

The CU has a 30mA RCD as main sw (probably installed under 16th regs), but all lights have their own switches, and all SO are switched ?

David