Induction hob tripping RCD

[Sparks]

Just checking it's not just me, but I think my new induction hob is faulty!!

It is connected to the RCD consumer unit (Because the cable that feeds it was feeding the oven, and was already connected there before you ask!)

When turning on the local isolator near the hob, the RCD trips. If I reset the RCD it will trip either immediately or after a few seconds If I isolate the hob (Via the double pole isolator, OR by the single pole MCB in the CU) the tripping stops.

I tried disconnecting it's earth (and keeping away from it!), and the tripping stopped If I measure the current from it's earth wire to earth, the RCD trips (no surprise there then!)

If I measure the current between its earth wire and Neutral, I am getting about 2.9A If I measure the voltage, it is about 120v

Now when it was first turned on (this afternoon), it tripped the RCD immediately, then after a few resets, it seemed to settle down - I even boiled some water on it to test it. I thought it must have been a bit damp or something (Although it has been sitting in a warm house for about two weeks!)

We went out and set the alarm as we would usually do, only to be called by the alarm about 10 minutes after leaving the house telling me my power had gone off (The hob had been powered up, but "off" for a good hour) Now the RCD refuses to stay on with the hob powered up - This is when I tested it.

Do I send it back, or connect it to the non RCD CU! (I personally would think 120v @ 2.9A leakage to earth is not too clever, so would think it is faulty, but would just like to see what others think)

...If I hadn't have connected it to an RCD, then this fault (assuming it is) would never have been noticed (Unless it got worse, and tried to pull more than 32A, tripping the MCB!) If the earth had then become disconnected somewhere between it and the CU, then someone touching it may have got a packet from it.

Is not RCD protecting things like this really as wise as people make out? (The same goes for an old dishwasher we used to own, the RCD kept tripping with this when the door was opened - turned out to be a neutral wire worn through, touching the chassis - again this would not have been highlighted without having an RCD...

Sparks (who wants to RCD everything possible, now more than ever!)

[Andy Wade]

That's a bit ambiguous - do you mean the current that flows through your meter with the meter leads connected between the neutral and earth terminals, with the earth connected? (Or if not, then what exactly?) A neutral-earth s/c current measurement like that isn't very helpful - what you need to do is to measure the current in the earth wire of the appliance. IOW connect the earth wire from the hob to the earth terminal of the cooker connection unit via your meter. Make sure the appliance metalwork is floating, i.e. isn't connected to earth through any other path. Now what earth leakage reading do you get? Anything more than a couple of milliamps is a cause for concern.

Strongly suggestive of a mains filter (RFI filter) and perfectly normal.

****

What (if anything) do the instructions for the hob say about earth leakage current and compatibility with RCDs? There is certainly no need for an appliance like this to be RCD protected, and there's no advantage in it being so (TT-earthed installations excepted). But that said, the leakage from a single healthy domestic appliance should not be anywhere near high enough to trip a 30 mA RCD.

[Sparks]

Yep, this is exactly what I did Disconnected the earth from the connection plate, then clipped one test lead to it, and the other the Neutral in the connection plate (as connecting it to the earth in the plate trips the RCD) I also tried clipping the test lead to the metal body of the hob and the neutral in the connection plate - both displayed the same results of 2.9A flowing (The test lead produces small sparks when connected too)

Hob is sitting on a granite worktop, and the parts touching this are plastic anyway so is isolated pretty well!

Nothing!

Because if the hob is faulty, as I expect it is, this fault would not have been identified as the leakage would not trip the 32A MCB - Say at some point the earth gets disconnected for whatever reason - the chassis of this hob will be ready to supply a user with 2.9A @ 120v - and as it is only leaking 2.9A no protection device would trip. Not only this, but I would be wasting nearly 350w of power 24/7! (That's 8.4KWh per day or 3068KWh per year (That's nearly £250 per year leakage!!)

Surely it has to be safer to provide more protection than is actually needed? (If not, please tell me why!!)

Sparks...

[Fred]

It is possible that there is sufficient current flowing due to a mains filter to trip the RCD. I've never been keen on suppressors between live and earth because of the resultant reactive current whereas the neutral current will be minimal.

[John Rumm]

Whoa - not the way to go. Remember that when you have your meter set to read current it looks like a short circuit across its probes (because it is meant to be placed in series with a circuit - not across it like when measuring voltage). If I understand what you have written correctly then you have just created an earth neutral short. A sure recipe for tripping a RCD in many cases. What sort of earthing do you have?

What you need to do is leave the live connected to the live, neutral to the neutral, and then disconnect the earth. Now connect one test lead to the earth connector on the appliance, and the other one to the earth wire you just disconnected. That will let you measure the current flowing to earth. You will probably need you meter on a lower range! You may still get a trip when you do this - but you should be able to get a reading for the leakage current. You could try turning off other RCD protected circuits at the CU for the duration of the test to desensitise the RCD (by removing other sources of imbalance/leakage) if required.

If it is faulty, then passing 2.9A to earth would be a rather odd fault. I expect you are simply measuring a large current flow between N and E as a result of a slight difference between your earth and neutral potentials. You would also not have been able to use the cooker to boil your pan of water unless your RCD was also seriously knackered as well since 2.9A should trip a 30mA threshold device rather rapidly.

(I take it it trips when you hit the test button?)

The 120V your read sounds like capacitive coupling. It is hard to think of a fault in a typical suppression circuit on the input of an appliance like this that would give you a 2.9A (i.e an 90 ish ohm resistance from live to earth) that would not result in something in the filter going bang in short order, or preventing the cooker from working.

Several reasons - you have already found one of them. Heating devices like this are often have high leakage currents (and we are talking milliamps here - not amps), which means you are eating into the leakage allowed by the RCD for all its protected circuits. The net result a greater likelihood of nuisance trips. RCDs are designed to protect you from electrocution in situations where there is a significant risk of you making contact with live - i.e. portable appliances, with the worst culprits being power tools used outside. For a permanently wired fixed appliance this will be very unlikely and the risk is minimal. Overcurrent protection and decent earthing gove you more than adequate protection (unless you are on a TT installation).

[Sparks]

Yea, I know! I wanted to see the current flowing out of the earth

If I understand what you have written correctly then

Nope, if I plasced the test leads on the earthe on the supply earth and supply ~Nuteral I would have done this

. A sure recipe for tripping

TN-S

That will trip the RCD, as the leakage is a lot more then 30mA

That will let you measure the current

As this method trips the RCD, I connected the probe the the Nuteral, as it is at the same potential as earth

You will probably need you meter on a lower range! You

It trips as soon as the meter touches the copper, so I cant get a reading quick enough!

You could try turning off other RCD

Not really applicabalw with my test - it is definatly dumping 2.9A to earth!

Well, sort of - the disconnected earth of the hob and the N of the supply.

I havent managed to measure the current when the RCD was staying on - so mabe when this was happening it wasnt leaking quite so much! The RCD is only about a year old - it was replaced as the pervious one went well over sensertive

Yep, and also when it was tested recently with an RCD tester :o)

[Tony Williams]

I've never understood why inline mains filters are not two series chokes, a cap between L-N, and a single cap from N to E.

[Peter Parry]

Probably not.

Induction hobs use high frequency and high current in the induction coils. The switching electronics produce a lot of RF interference which needs to be filtered to stop it getting back into the mains supply.

A whole house RCD isn't fussed about where leakage is coming from and if you have a number of noise filtered devices (such as computers, printers, fax machines)on other circuits then each can leak about

3.5mA and it is likely that the RCD is already on the point of tripping and the hob filter has just pushed it over the edge. Noise filter leakage usually occurs whether the device is switched on (using its own power switch) or not.

The current reading is irrelevant as has been pointed out.

Which is exactly as you would expect from a noise filtered device.

It isn't there. Your measurement ignored the neutral/earth voltage difference. You would get similar results from any device in the house.

Connecting lots of things to a single 30mA RCD is a recipe for nuisance tripping. You seem to have somewhat of an attachment to RCD's so if you really want one connect the hob to a non-RCD circuit and put a dedicated RCD on the hob alone.

Yes - whole house RCD's have killed more people than they have saved. Increasing use of electronic devices and the increased need for RF filtering means many houses now qualify as high protective conductor current environments and should meet Sect 607 requirement.

covers Sect 607 well).

If you feel you must then use lots of individual RCD's at each device you wish to protect.

[Fred]

2.9A reactive current sounds far too much. In reality the earth will be close to neutral potential so the measurement is a fair one in this case. You wouldn't allow the earth to run at 120V would you? BTW I hope these are AC measurements and not a DC ones!

I am certain that new standards would forbid such a poor power-factor or such a reactive current to flow even from live to neutral.

I would ask the technical department at the manufacturer to reflect on the measurements before doing anything further.

[John Rumm]

In which case can you move the supply to the non RCD side of the CU and repeat the test?

Your Neutral is very unlikely to be the *same* potential as earth - close certainly, but given they are both going to be low impedance paths there is a good chance a substantial current will be able to flow between them if connected.

So that measurement is taken with the hob effectively not earthed at all (i.e. the earth wire from the supply cable was not connected to anything)?

If you diconnect the hob and measure the resistance between L & E and E & N and L & N (with all the switches/controls on the hob turned off), what readings do you get?

[Sparks]

Maybe you are miss understanding me, it is hard to explain everything without trying pages and pages!...

But this is happening with the hob on standby.

Sorry, but you must be missing the point here! I am NOT measuring between a connected Neutral and a connected earth, simply a connected Neutral and the DISCONNECTED earth of the hob (Or the chassis of the hob, with the Earth wire DISCONNECTED)

I would hope I wouldn't (as explained above)

So I will then get the 2.9A leaking to earth, 24/7 - Do I really want this?

I never asked about whole house RCD's as I don't have one. (I have a non RCD CU for things like the freezer, fridge, alarm, sub-main feeds to further CU's in outbuildings (These of course have RCD protection locally) and a separate RCD for the lighting circuit in the house, and another one for socket's etc.

Sparks

[Sparks]

I havent tried this, as I felt it was pointless due to the results of mu tests I can do this is it is really necassary, however it will be a bit of a mare to do!

Correct - The hob was placed on the granite worktop, with only plastic parts of the hob touching this worktop, so the appliance was not earthed for this test.

Not Quite as simple as this, as it has electronic touch controls, not knobs or switched - however the readings are: -

L-E - Starts at about 11Meg - then rapidly climbs to infinity N-E - Ditto above L-N - Starts at about 4.5meg, and settles at about 1.4meg

Sparks...

[Peter Parry]

It is, I'm not altogether sure what is being measured here but a current reading from chassis earth to mains earth is what is needed. I'd expect it to be a few mA.

He is measuring voltage from the floating hob chassis with the earth disconnected. The noise filter capacitors will put the chassis at half mains voltage in this situation.

That's why you treat mains circuits with high protective circuit currents carefully and take particular care with earth integrity.

[Peter Parry]

The filter is on the input - it is in circuit all the time unless the hob isolator is open.

If you were really getting 2.9A flowing at 120V there is 350W being dissipated somewhere in the hob all the time and you would feel the heat generated without any problem at all.

The fact that the neutral to isolated chassis voltage is 120V suggests the noise filter is serviceable and there is no marked extraneous conduction path. The RCD is not always tripping immediately which means there certainly isn't 2.9A or anything remotely like it flowing all the time.

Can you connect the hob to a non-RCD circuit temporarily and measure the current between the hob earth and mains earth?

What is the voltage between neutral and earth at the hob connector?

Why would you have this? The hobs dedicated RCD would trip immediately in this situation.

[Michael Chare]

What brand of Hob?

I have a Bosch induction hob which failed when new. When the Bosch engineer was working on the upside down hob to repair it he got a shock touching some part you would not expect to be live. We measured the voltage with his digital voltmeter and got a reading of about 80v (or so). In normal operation the metal bottom of the hob would have earthed the internal part that he touched. Not having an RCD I live in happy ignorance of any currents that may be leaking to earth.

[Peter Parry]

All induction hobs (and any other piece of equipment using the usual noise filters) will put "earthed" components at half mains potential if the earth is disconnected. This is not a fault but a perfectly normal characteristic (and why Sect 607 earthing should be considered in houses). The current will not be enough to kill or directly cause injury but can cause indirect harm by causing you to fall off ladders, drop things etc.

[Christopher Key]

There've been quite a few replies explaining that this measurement isn't valid, or at least is irrelevant to the problem, but I am note entirely clear on why. My understanding of the test that you are performing is as follows:

Appliance L connected to Supply L Appliance N conencted to Supply N Appliance E - not connected Supply E - not connected

You then place the multimeter from the appliance earth to supply neutral, and it read 2.9A. To me, this clearly doesn't sound healthy. In a healthy appliance, the earth should isolated from both live and neutral (with the exception of a high impedance link via the noise suppression caps), and should hence not be able to supply any current to neutral.

The point that was being made is that if you place a multimeter between the supply earth and neutral then significant current will flow. This is clearly true, as N and E are held different potentials via a relatively low impedance, depending on the earthing system. However, in Sparks' case, he is not measuring current from supply earth to neutral, but from the appliance earth to neutral, and the appliance earth should be isolated from any potentials as he has disconnected the earth terminal. It is possible that some form of supplementary bonding is in place, and introducing a potential on the appliance earth, but this seems unlikely because his RCD ceased to trip when the earth terminal was disonnected. Also, the fact that the RCD doesn't trip, even when the multimeter is showing 2.9A flowing to neutral, suggests that this current is being sourced from the RCD protected supply.

If my understanding of the tests that have been performed is correct, then this does sound clearly like a faulty device.

Chris Key

[Sparks]

At last, someone who can see my point :o)

I eventually managed to get through to someone at Siemens who has a clue, and it is going back!

[Peter Parry]

In most appliances I would agree but induction hobs have a heap of electronics in there to produce the high frequency high current needed to operate. I'm not altogether convinced the test is valid hence the suggestion that an earth-earth test be carried out on a non-RCD circuit.

There are a number of inconsistencies:-

If the leakage current really was consistently a few amps the RCD would always trip immediately. It isn't. It is tripping after some time and somewhat unpredictably - this points to a situation where the total asymmetry in the house is hovering about the minimum trip current.

The voltage is about half mains - this is exactly what you would expect from a noise filtered circuit with a floating earth - any leakage path would alter this reading.

I'd expect a few mA leakage through the filter.

Except that the results of the tests and the reported fault symptoms are mutually exclusive. With a permanent leakage of this level the RCD would always trip immediately but apparently it isn't. The voltage reading is also inconsistent with a single fault capable of supplying this current.

[Christopher Key]

I can certainly imagine the the high frequencies being leaked could confuse a DMM, but can't see why the results would be any different if the leakage was measured to earth and not neutral. Certainly no harm in trying if it's possible though.

I take your points onboard, but still maintain my original argument that the multimeter was correctly connected to measure any leakage to earth. Whether it was actually doing that, or reporting spurious values as a result of induced noise, or an odd waveform is another matter. Would be interesting to see what happens with a moving coil meter.

As you say, there is no way it could be leaking 3A to earth consistently, and only occasionally trip the RCD. It is conceivable that it is an intermittent fault however, as reading the OP, I got the impression that typically the RCD tripped instantly, but occasionally was ok for a while. This could be consistent with an intermittent short internally, that was open when the 120V was read, and whilst the RCD wasn't tripping. That does seem a little far fetched though.

There are combinations of whole house faults that could cause the behaviour described, but they are so contrived that I doubt they could be created even if one tried!

It seems like the hob is being replaced however, so I shall be interested to see the behaviour of the replacement.

Chris Key