RCD didn't!

[John]

Switched on the power to the pump in the garden pond now the ice has thawed but nothing happened. Put hand into water to lift pump out to see if was blocked and got an electric shock (strong tingle). Switched off power and recovered pipe to find the mains cable had become severed under the water - whether this was because the ice had cut through it or one of the squirrels had chewed it whilst it was above the level of the ice.... Anyway the RCD was not worried - ooh err! Will double check the earth connection to the pump when I replace the cable.

[Tim Watts]

Have you tried the test button on the RCD?

And for something like this (garden supply) have you been testing the RCD at least once a month?

It's possible that the leakage current may have remained below 30mA but I'd be surprised in a pond with all the yummy conductive ions in the water and the potential to find a good earth, even if capacitively through a large surface area pond liner.

How big is the pond?

[tony sayer]

In article , John scribeth thus

Any deceased Squirrels around to support that theory;?...

I'd more check the RCD is working OK..

[Grimly Curmudgeon]

We were somewhere around Barstow, on the edge of the desert, when the drugs began to take hold. I remember John saying something like:

They do fail, you know. I now tend to regard them as a long-life consumable.

[Pete Zahut]

Our pump runs 24/7 365 days a year. Ice at the back end of the pond was about 3" thick but the area around the waterfall was clear and kept the pond nicely aerated.

[Ronald Raygun]

The squirrel will have been removed from the scene by a fox, delighted to come across a freshly par-cooked one.

[www.GymRatZ.co.uk]

My "project pond" missed completion before winter and our 6 carp average

18 inches each are wintering in an 8' paddling pool. Pump and air stone left running as the only sure way to prevent complete freezing. Slab of Celotex thrown on top for good measure. Recent thaw and all the boys are scuttling about as normal. Leave the pump running and keep the fish (and pond and pump) alive.

;¬)

[The Natural Philosopher]

remember the earth to the pump is not an issue for RCD. What is, is the current the goes down the live that doesn't come back down the nootral.

Possibly the tingle was well sub 30mA.

[Tim Watts]

I would end to agree - but if the water was "live" it would almost have certainly been leaking plenty through the rest of the pond.

[Steve Firth]

A few years ago we had a nasty following a visit by the electricity supplier to the house. They wanted to change the meter tails and for some reason in the process their electrician decided to remove the cover from the RCD (no need to do this that I can see). What we didn't know was that when he reassembled the RCD he managed to do so with the cover fouling the RCD switch lever slightly. When the RCD did trip the lever could not move and the circuit remained live. Fortunately the short that caused the trip was as a result of poor practice by a "professional" electrician not because of a child or wife being the short circuit. The MCB tripped and prevented any further damage.

The fact that the RCB could be obstructed like that puzzled me. I thought that even if the switch were held in the "ON" position the RCB would still trip. Clearly not for the MK unit that we have. Slackening off the single screw that held the cover in place, repositioning the cover and testing many, many times showed that the problem had been "fixed" but I'm still not that inclined to trust the damn thing.

[A.Lee]

Or, more likely, flowing back to neutral, and heating the water between the 2 contacts, just like when an immersion heater corrodes through.

Alan.

[John]

In article , tony sayer writes

It is a fake stream - pumps from pond up to top, down over a couple of waterfalls (no fish). I had switched it off before it froze and only turned it back on today (so no dead squirrels!). The cable had been cut/chewed through and all three conductors were in the water. I was surprised that the water - at least 0.5m from the cut end was live enough to give a strong tingle without something having tripped. Anyway, I will investigate further tomorrow and test the RCD (which was only replaced last autumn)

John

[ARWadsworth]

It might not have been the physical restriction on the ON switch that stopped the RCD tripping.

I have seen RCDs fail to trip when they are under a physical stress caused by missed placed lids (not actually touching the ON switch) and again in a botched DIY attempt to replace the RCD with another manufacturers RCD and they had forced the new RCD onto the bus bar as it was not a good fit.

Of course the dozy sod should not have been messing with your RCD.

[John]

In article , John writes

RCD Test button tripped it OK

[Steve Firth]

Hmm, interesting, thanks. I wasn't aware that could happen. Presumably it's mechanical distortion of the unit that does it?

Indeed.I'm still not sure if he was trying to drum up extra work. He kept mentioning a survey and rewiring to my wife - I was out at work at the time - but he admitted that the installation was relatively new and there was nothing wrong with the bits he could see.

[ARWadsworth]

As far as I can tell, yes. Now this is something that I have only come across a few times out of all the RCDs I have tested. It is worth noting that the RCDs did operate properly when not distorted.

In fact one of the times I saw it there was no lid on the CU. The wall was very uneven and some numpty (apprentice) decided to screw the CU to the wall without bothering to use washers to level the back of the CU. He over tightenend the screws putting a lot of twist onto the bus bar, RCD etc. There was certainly enough twist in the plastic CU to stop the lid fitting!

As has been mentioned in other threads, RCDs sometimes do not operate if they are left in the on position for a long time. A manual OFF then ON clears the sticky behaviour of the RCD (not that you could do that with your ill fitted lid)

Happy New Year

-- Adam

[ARWadsworth]

But would it leak through a plastic liner? (not intended to be a pun)

[Steve Firth]

And to you and yours.

[Tim Watts]

I reckon the capacitive coupling would leak a fair bit. Liners are thin and pond surface areas could be quite a few m2.

Quick envelope calc with google factoids:

Butyl rubber has a dielectric constant of about 2.4 If we assume a 1mm thick liner, 10m2 area total area (walls and base) as a parallel plate capacitor, that gives a capacitance of about 0.2uF which gives a reactance at 50Hz of around 16k Ohms

So the RMS current at 240V would be 15mA

(Corrections welcome!).

10m2 would be a pond 0.5m deep and a bit over 2x2m or variations thereof.

Google suggests that butyl pond liners are in the range of 0.75 - 1.00mm

So we're half way to tripping the RCD on reactance alone with a decent sized pond - not allowing for other L-E leakage on that circuit and not allowing for resistive leakage over wet pond edges to wet ground.

=====

So a reasonable conclusion that the RCD *might* not trip even on good wire contact with the water - but it could easily trip (more leakage, thinner liner, bigger/deeper pond).

Mind you - I would have thought the OP sticking his fingers in would have pushed it past the tripping threshold...

Debate welcome!

Cheers

Tim

[ARWadsworth]

Well I am possibly the only person to have a customer who worked out that the deeper he placed his arms into the water the bigger the electric tingle!

That was not an RCD protected circuit. It is now.