Online assistance for electrical trade people Login  |  Register  |   Forgot Password
Assistance for electrical trade people




Click here to send Ron a pdf document for publication on this Topic

Documents must be less than 200k in pdf format

Posted By Topic: RCD Tripping.

Oct 04 2018 06:58

After 40 years in this game I thought I could solve most problems, but this one has me beat.

The scenario:
A Motel unit having a kitchen refit.
2 double plugs on the bench unit.
1 four plate ceramic hob cooktop.

The plug circuits are fed from an RCD on the sub board.
The hob is fed directly from an independant MCB via the main switch with a new cable solely for that.

The problem.
Every time one particular element on the hob is switched on the RCD for the plug circuits trips. The other three elements run up with no issue, as does the fourth after its tripped the plugs RCD

Its the same element every time, and it does it on a replacement hob as well, on the same element.

The hob is NOT on the same circuit as the RCD, and doesnt share the RCD neutral bar.

There are No cross overs on the Neutral bar, every circuit fed from the RCD has its neutral on the correct bar, and the unprotected circuits are on the other bar.

I have ruled out a faulty batch of hobs by dismantling and checking each one, and with an independant earth I have ruled out any transpositions, which was unlikely given that it only happens on one element and not all four.

Any ideas??


Oct 04 2018 09:01

So you switch on an element on the hob and the RCD for a different circuit trips. That is a bit unusual, the usual suspect being a crossed wire which you've ruled out, and it cant be anyway as its just one element that triggers the problem.

First two things I'd do is disconnect the plug circuit from its MCB, all three wires, leaving the double sockets in place,, and the same with the hob circuit at both ends, and megger every combination of cores within each TPS and between TPSs. Rule out the obvious.

The second is I'd throw a sheet of plastic or a tarp on the floor, relocate the hob onto that, and see if the problem persists. If it goes away then there is something about the location (coupling?) or structural connections causing the problem.

The third is I'd temporarily supply the hob direct from a MCB no RCD. If it goes away there's something about the RCDs or their wiring in the board.


Oct 04 2018 09:04

And... while you're meggering, check every core to earth as well.

Oct 04 2018 09:07

Had something similar recently. Turned out the RCD tripping had a faulty ballast on the cct which wasn’t showing up on an insulation test. But every time another non RCD circuit was switched on the RCD would trip. Took the other sparky 12 months to find it.
I’ve also had a TPS that a nail went through and nicked the cores. Even 1000V insulation test wouldn’t break through the glue on the nail. But 3 seconds after turning on the cct breaker would trip. Only found by isolating the circuit into sections. You may need to start isolating sections of the cct with the RCD.
There’s some interesting things that happen out there.

Oct 04 2018 09:12

I may have misread the question; are the sockets fed from a sub board and the hob from the main board?

Oct 04 2018 09:53

Can you temporary fit an appliance lead to the Hob and power it from the RCD protected outlet, check each element one at a time (be aware of power draw here).
Is it the biggest element that's causing the problem.
Can you run a new piece of TPS (again temporally) from the MCB for the hob straight to the hob and re test, i.e remove the current wiring from the situation?

Oct 04 2018 12:59

Ok, thanks for the tips.

Dbuckly, The hob and the plugs are on the same sub board.

I’ve already had the job, two of them in fact, running off a flex and mcb with no issues, but it’s worth a shot running it through an rcd. Although that still won’t explain why one is tripping on an unrelated circuit.

It’s one of the small elements tripping the RCD so it’s not load related.

A nail through the cable is possible, but unlikely given that it’s only one element causing the trip, however direct connecting to the board has merit and worth a shot.

The bench is simple custom wood and Formica with no metal nearby, the sink is right at the other end, so I don’t really think environment is the issue. That said, anything is worth a shot at this point.

There are no flouro lights in the building, so no ballasts to contend with, and in any event there’s no RCD’s on the lighting circuits anyway. Predates that rule.

There’s nothing at all plugged into the rcd circuits yet, so no loads at all for the rcd to see.


Oct 04 2018 14:18

Leakage Current Clamp to measure the leakage?

Test the RCD with a "ramp test" to check the tripping current point.

Oct 04 2018 17:02

Makes No sense That one element on the cooktop could cause this and not the others ?
Is it an Induction hob or just a standard old normal element Hob?
Couple of things to check if you have not already would be that the Line end phase and neutral to the RCD are on the "same" end , and that the RCD is Not a polarity sensitive type i.e. you can have Line or Load on either end.
Wrong connections can cause issues with the trip circuit in some RCd's .
That said it still doesn't explain why it only trips with the one element.
You have said there is nothing connected on the outlets so disconnecting the load side of the RCD , does it still Trip ? if so there has to be a wiring issue of some type with the RCD.
Once again the weird bit is why just with one element?

Oct 04 2018 17:55

Not an induction hob, just a standard old smoothtop with wire wound elements.

Didnt have much time on this today, but I tried a different RCD, not polarity sensitive, and neither was the original, but nothing changed.

I left in disgust after that, but late this afternoon i wondered if somehow there is two feeds to the plug circuits.

I havent checked for this yet, but if the plugs are looped together, which they will be, it may be possible that there's a feed at each end. One from the RCD and one from an unprotected MCB, and when the hob is turned on it could be 'back feeding' the RCD on the load side ???

Still doesnt explain why its only one element tripping the RCD though...

Time to pull everything apart and start again I feel.

Oct 05 2018 15:15

If the two cables are in parallel, could it be induced currents upsettings the RCD?

Oct 11 2018 20:25

Hi, we had a similar situation. Using the power outlet on a non-rcd oven tripped the rcd on the kitchen power outlets. After a long search we found that the interior cladding of the flat was metal, the old flush boxes were metal and old power outlets with exposed terminals and build up of crap in box, combined with old stove with a bit leakage and somehow some current from the stove was tracking to the power outlet and tripping the RCD. Fitted new power outlets on insulated mount blocks, new stove isolator and problem gone.
Don’t know if that will be helpful for you or not, good luck.

Oct 12 2018 08:34

Thinking about it, is it a metal bench top, if sonhas the earth conductor for it been run from a power outlet and not back to the switchboard? If you have a element with leakage to earth, which I say you do, then it could been tracking back via this.

Oct 12 2018 08:34

Thinking about it, is it a metal bench top, if sonhas the earth conductor for it been run from a power outlet and not back to the switchboard? If you have a element with leakage to earth, which I say you do, then it could been tracking back via this.

Oct 12 2018 22:31

Sorry, should've updated this days ago. I found the issue. Turns out that the adjacent plug circuits had two feeds. Essentially a ring main circuit, with one end fed from the RCD and the other end from the unprotected side, and when the hob was turned on it immediately created an imbalance across the RCD and it tripped.

It also meant that all the pugs were still live after the RCD had tripped, and the bizarre thing is that it was only one element on the hob that was tripping it.

Oct 13 2018 07:35

So when you changed the RCD the feed side was still live?

Oct 13 2018 07:48

After some thought

....You would have turned the main switch off.

Oct 13 2018 12:58

Uhuh. After 40 years I think ive learned how to stay alive quite successfully.

Oct 13 2018 16:43

Once you had found the wiring error and corrected it did you fully test the operation of the RCD?

It was reported in Australia in VIC that when an RCD has connected so that the a.c. supply input was connected to the LOAD terminals of an RCD, the RCD sensing circuit was destroyed and failed to operate correctly.

Noting that only some makes of RCD are destroyed when the incorrect supply connection are used.

In Australia in VIC you have to use selected RCDs only which have then tested to be uneffected by incorrect external supply connections.

By the way the RCDs generally used in Australia are type AC RCDs, a type that should NOT be in general use in NZ because ESR 2010 does NOT permitted the use of Type AC RCDs for personal protection, as they don't provide Low level pulsating DC or continuous DC residual currents protection caused by most electronic based equipment, such as LED drivers and speed controlled appliances.

Oct 13 2018 17:16

I'd guess the failure cause in that situation would be the spike from sudden disconnect when the RCD operated. Because the RCD was reversed then it's own supply was incorrectly fed from the switched end and therefore had to soak up the inductive spike coming back from downstream.

Symon's situation was still low impedance and still live on both ends of the RCD. Not much to spike it with. Probably didn't even have any contact arc.

Test it, sure. Worried about subtle damage, don't be.


Oct 13 2018 17:24

Did that AU report attempt to identify the device failure mode, from an engineering view?


Oct 13 2018 20:21

evanh Oct 13 2018 17:24

Your comment
Did that AU report attempt to identify the device failure mode, from an engineering view?

My comment
Found is result of an electrical accident that was reported to the VIC electricity regulator.

It is my view that the adopted solution was "a bit over the top" for what was an incorrectly installed (connected) RCD, the installer failed to follow the RCD manfacturer's instructions or the RCD supply connection labelling.


Oct 15 2018 08:29

Good to have found it, did you have an RCD tester? This should have picked this up straight away.

Oct 15 2018 17:24

Given that we know a large number of electricians still don't do the mandated tests; why do we seem so eager to assume that an existing installation was actually correct when first installed?
When this sort of gremlin strikes, should apply Sherlock Holmes' principle: having eliminated the possible causes of the problem, the actual cause must be "impossible".

Oct 16 2018 21:28

What are plug circuits? Is that where plugs hang out of the wall? Or is it where circuits are plugged in :/

Going back to the OP, he does say a new refit, which indicates new electrical work, and not some fault that has just popped up one random day with the installation in service.