RCCB Tripping in Case of Phase to Ground Fault on Upstream Network

Thank you for your reply.

As neutral is grounded at load as well, won't fault current(or a component of it) find a parallel path through neutral ground to source through neutral line there by causing tripping?

Thanking you beforehand for your impending reply..

Did you read #1?

Connection between earth and neutral is subject to a code, which can be found in Wikipedia on the page about earthing systems.

If in doubt, consult a qualified Electrician, who will warrant to you with a document that the installation is installed and tested to the applicable national code when that individual left it. If you don't know what you're doing then you really shouldn't be bugging about with it.

Thank you for your patient reply..

Sorry to bother you again, Mr. Crabtree,

To conclude, if neutral gets earthed downstream of RCCB, then it will get tripped for a phase to ground fault in between source and load, right?

..."RCDs operate by measuring the current balance between two conductors using a differential current transformer. This measures the difference between current flowing through the live conductor and that returning through the neutral conductor. If these do not sum to zero, there is a leakage of current to somewhere else (to earth/ground or to another circuit), and the device will open its contacts. Operation does not require a fault current to return through the earth wire in the installation; the trip will operate just as well if the return path is through plumbing, contact with the ground or any other current path. Automatic disconnection and a measure of shock protection is therefore still provided even if the earth wiring of the installation is damaged or incomplete."...

https://en.wikipedia.org/wiki/Residual-current_device

http://www.studyelectrical.com/2015/05/residual-current-circuit-breakers-elcb-rccb-rcbo-working-principle.html

No! Circuit Breakers are installed on the HOT or source side of the power line.

If your fault occurs upstream, the voltage goes to zero and then the circuit is broken by some other upstream device. Consequently, at the time of the upstream fault you get nothing downstream and it is not possible for your RCCB to trip.

Just like a river. If it is dammed upstream, you get nothing! (for a while anyway...)

Thank you for your reply.

What if the fault path has high impedance and so fault current is not large enough for an upstream overcurrent device to operate?

Then the upstream device will not operate.

If there is only one connection between neutral and earth, as is standard for most earthing systems, an RCCB may well pick up any other extraneous connection. They are a bugger to find sometimes and removing all of them is a Good Thing.

Did you read that page on earthing systems at all?

I have read earthing system of Wikipedia. It explains a TT system with a fault on downstream of RCD, as in if fault path between source and accidentally energized objects has low impedance, it will cause overcurrent device to open otherwise RCD will open.

My question is about an upstream ground fault(w.r.to RCD) on TT system i. e. a line to ground fault between two earthings of TT system. For simplicity, lets assume that there are no upstream (w. r. to fault location) overcurrent devices in the system(even if they exist, they will be selectively coordinated) . As no metallic conductor is present in that fault, the fault path has high impedance.

Now read #2 & #5 and tell me whether RCD at load trips or not? Yes or No?

Look. A RCCD of 30mA trip setting on a 230V supply will only support a fault of 6.9VA on the live. On the neutral, it might be as low as 0.1VA. You can't do a lot with that, apart from correcting it.

However, rest assured that any neutral to earth fault will be demonstrated eventually by the RCCD.

So stop messing about on here, and get busy chasing out those faults before someone gets hurt/killed. It's urgent. Being on here is not.

You wrote:-

To conclude, if neutral gets earthed downstream of RCCB, then it will get tripped for a phase to ground fault in between source and load, right?

Of course......that is usually the requirement!! The underlined part is simply not needed.....nor is a load either!!!

Have you gone through #5?

Why do you all presume that neutral grounding at downstream of RCD is planned?

Neutral is not grounded at downstream of RCD as a standard practice. It's not premeditated, may be accidental.

Now tell me the fault path for said fault and RCD status.

A) Because it is stated in the original post. And it's folly.

The RCD only operates when there is a difference in current between the live and the neutral passing through it in excess of its trip setting. As the only other circuit path is to earth, either via the protective earth conductor at one extreme or via a human being at the other, it assumes the worst and trips. That is what it is for.

Your English is difficult to understand.

As written, it appears to be (my opinion only) that you do not understand what needs to be done.

I am fairly confident that having ground and neutral linked twice is completely wrong, but as to whether it would cause tripping in itself, I am personally unsure, but possibly......

You are also unclear in your posts as to the position of the RCD in relation to both ground/neutral links.

Make a clear diagram in jpg format, place it here and maybe we can help better....

What's there not to understand, you know technical terms of electrical engineering, right?

I have explained a scenario more than once in all my posts. All of you keep thinking that it(Neutral being grounded twice) is wrong. Yeah, it's wrong. FYI, that's why it's called "Fault". I am trying to understand the system under that fault. Stop being a stickler for code and start imagining.

If you read all of my posts carefully and try to imagine even a little, you would have understood the problem by now.

I am not changing location of RCD or any other element of system any where. I just assumed that you had the basic knowledge of the general distribution network(from source to load). A simple power system actually.

Thank you for trying anyway.....

Its still your problem then and if you are as half as clever as you feel you are, why did you ever come here???????

The problem is simple and several here have told you why.....but obviously you did not understand....!!!

I'm off!!

Byeeeeeeeeeeeeeee

I was under the impression that the forum consists of all knowledgeable people, but you proved me wrong... ��

Thank you anyway..

If anyone understands the problem and thinks that they have a solution which is as claimed "SIMPLE"

Please enlighten me by telling the fault path of the said fault condition and RCD status.

Many above already have.

Let me try again. Start with the device closed and no fault. The device's status is "normal". A fault occurs. The new fault path is from either the live or the neutral conductor, via the fault, into the earth conductor, which might actually be local earth via a person, and back to the supply that way. The fault current causes a mismatch between the live conductor current and the neutral conductor current that, once it exceeds the device's trip setting, causes the device to open and disconnect the fault, dropping the fault current to zero. The device's status is now "tripped".

Are we done yet, Mildred? Only I've got a couple of British Airmen to get back to Blighty...

Is your exact fault location upstream of RCD that got tripped?

I haven't got a fault.

If a fault occurs upstream of an RCD then a device upstream of the fault may well operate so as to disconnect it. As to whether the RCD operates as well, then that depends upon whether it is active, i.e. sustained closed by the supply to it, or passive, i.e. the fault current operates it. Both types are available and, as I can't see your installation and it sounds as though I shouldn't be going near it, only you can know the answer.

If a live or a neutral conductor breaks upstream of a passive RCD, then the device will not operate because the currents in both the live and the neutral fall to zero simultaneously. However, if it is an active device, it will need to be reset manually once power has been restored.

If the fault is large enough to exceed the trip setting of a device further upstream, then that device will operate to disconnect it, irrespective of its actual impedance.

Are we done yet?

Come on, man. Stop nitpicking, when i say your fault, i mean the fault you have considered.

Consider a simple power system with TT earthing system containing source (can be secondary of supply transformer) and a single phase load (one house) being supplied by said transformer and located at farthest possible distance. Ignore everything else including persons for ADS. Passive RCCB is considered for discussion here.

Now if neutral gets earthed accidentally in an equipment at load, will RCCB trip certainly given that there exists high impedance from that point (second T of TT system) to source earthing (first T of TT system) via the earth? Is it a certain event considering the entire system?

On the other hand, in your trailing answer, you have considered everything except accidental neutral grounding at load (downstream of RCCB). Now if a line to ground fault occurs at a location between source and load and neutral gets earthed subsequently in the said manner, that fault has two parallel paths.

1. Source to fault location (via phase) to the earth and back to source via first T of TT system. The magnitude of the current depends up on the exact location of fault.

2. Source to fault location (via phase) to the earth to second T of TT system and back to source via neutral of the system. The magnitude of the current depends up on the exact location of fault.

If the said fault location is closer to source, then RCCB at load won't get tripped. But if the said fault location is closer to load then RCCB will trip isolating the fault from load, right?

I'm getting the feeling that you ought not to be messing about with stuff you don't understand.

We need a diagram....

If there is a fault at A-B, then fault current will go to earth at both the earth points. Theoretically, with voltage A to B =0, iz2 will not be matched by a current in the live to the house.

But it is important that iz2 does not pass through the coil of the RCCB, so it will not trip.

Note that everything electrical in the house is a) insulated from ground and often b) surrounded by metal which is connected to the house wiring earth terminal at the house incoming point via the circuit protective conductor CPC. Incoming metal water/fuel gas pipes are bonded to this house wiring earth terminal as well as internal metal pipes (which may not be earthed automatically because supply water /gas pipes are often plastic now for new building).

You are justified to worry because there is capacitance to earth from both live and neutral. While the capacitances are mostly equal, the voltage of live to earth is far higher than neutral.

When a breaker is closed to energise the load, the charging currents are not equal, current in the live is higher - because its voltage to earth is much higher than neutral to earth. This difference is seen by the CT of the RCCB and the current/time can be enough to trip an RCCB.

This cannot be a problem for the normal house; because systems with multiple earths, one at each house, as well as at the substation are common (combined neutral and earth up to house also).

However, it is a problem for circuits with a lot of electronic equipment which have noise filters having capacitors from live to the CPC (often .001 to .005 microfarad each equipment - so 40 equipments might be 0.2 microfarad). The capacitor current from the live goes to the CPC , not the neutral - it causes an unbalance in the RCCB CT which may cause a trip.

A capacitance of 0.2 microfarad at 230 V AC, 50 Hz, draws about 14 mA - note that the typical "30 mA" RCD is function tested at 15 mA to ensure it will NOT trip as well as at 30 mA to ensure it does trip. Trips are likely to occur when the circuit is turned-on, because the voltage may be 330V peak at this instant and a "DC" pulse occurs in the live, with a peak current of amps.

You may have noted that RCCBs at the incoming point of an installation are recommended to be at 100 mA or more. This is probably because the capacitance live-cpc of 1000 metres of cable will be about 0.2 microfarad.

67model

Thank you for your reply..

I agree with your answer. However, it is not necessary that voltage between A-B(point B is on the earth) is zero. Earth potential is known to rise locally especially in case of earth faults.

But if neutral gets earthed downstream of RCCB, it will get tripped for the fault considered (i.e. earth fault between source and load) even if parasitic capacitance is ignored, right?

Because the component of current iz2 of the diagram (if sufficient) will flow through grounding conductor at load to grounded neutral and back to source(via neutral) thereby causing RCCB to trip.

You are right that a neutral to earth fault downstream of an RCCB could trip that RCCB - given enough current. The ring CT around live and neutral does not care if current gets "lost" from neutral or live.

The "hot" wire gets the attention, but there is potential between neutral and earth due to the load current flowing in the neutral.

In a sub-circuit, which may have up to 10% volt drop, the neutral to earth voltage could be e.g. 5% x 230V = 11.5 V.

On the smallest circuit, the live-earth/neutral-earth loop resistance may be about 15 ohms, but even this gives 11.5/15 = 0.77 amp for a dead short. This is above the common 30 mA RCD trip current.

I suggest you "prove a point" by putting a 230V 60 watt filament lamp between neutral and earth of a socket outlet - it will probably trip a 30mA RCCB (it does depend on voltage drop in neutral due to load - you can measure neutral-earth volts at every socket to find the one with most volts

A neutral to earth short is far less likely than live to earth.

This is because neutral and everything connected to it is insulated to the same level as the live. The reason for this was "brought to my awareness" earlier this year.

The electricity supplier decided to change the kWh meter for my church. The cables to and from the meter were changed also. Not long after, a live-earth fault in a socket occurred.

Everything went off. Investigation found the neutral was 245V to earth, even at the isolating switch fed from the meter. This was outside our responsibility zone, meter connections are security sealed. A colleague found that nothing would be done till after Easter - unless the "safety" man was called. So he was telephoned and came, rubber gloves, boots, a mighty powerful torch and all.

Soon after opening up the incoming cable head, he said "I think I can see your problem!" delved in and brandished a cable end. This cable end should have been clamped tight in its terminal! In fact, both of the two clamp screws of the terminal were tight, but had not engaged the cable - the cable size was small for the terminal.

Someone did not pull hard on the cable to be sure it was gripped.

He proceeded to demonstrate how the job should have been done - wind a spiral of solid copper wire around the cable copper end so it just fitted the terminal.

If you have a low resistance to earth on your neutral, to find the cause is difficult, because many items, like timeclocks, are connected neutral to live - the latest electronic ones I fitted had their clock before their switch (on line side). If you have double pole RCBOs on each sub-circuit life is easier!

If you are lucky, you have a two pole main isolator [live & neutral], otherwise you may have to disconnect the main neutral cable.

Thank you for the explanation.

I felt that path of fault current component iz2 i.e. circuit (live wire) to earth and back to circuit (neutral wire) was counter-intuitive when i first stumbled upon this problem.Now, i find the same reasonable with your two answers.

I will try to measure the neutral-earth volts at various points in the path for better understanding.

Thank you again for good answer.

My measurement (DC) of a 60W 240V filament bulb found it drew 47 mA at 4.9 volts.

I would appreciate your reply with your test results.

A modern trend is to have overvoltage surge protection devices at the incomer of a building or maybe at a subcircuit. These are connected line-neutral, line-earth, neutral-earth - so give another path for "not-expected" "earth" currents to trip RCDs. Particularly, if someone has gone down the "separate earth rods for power and instrumentation" path.

Instrumentation (and sound/HiFi systems) frequently raises debates about multiple earth paths you wish you did not have or did not know about until you hit a problem (plenty in CR4)!