Investigating a faulty earthing in our house

[artis]

Well @Wrichik Basu , no offense but the pictures you provided show a rather dirty fuse box and the neutral and earth connections do seem oxidized and rusted outright, that can be a problem or part of it, but to know one would need to measure the contact resistance and visually inspect them thoroughly.
I recall one late friend of mine who was an electrician once said to me:" There are only 3 problems in all of electronics, namely
1) A contact where there should be no contact
2) No contact where one should be
and
3) Searching in the wrong place

Apart from the possible ideas already said here one additional idea came to mind.
If I understood you correctly you have a earth wire in each outlet of the house besides the two wires of neutral and live.
So a 3 pin socket essentially. Now apart from the good suggestion that @Baluncore already said in post #5, my additional idea is this.
Clearly apart from bad ground appliances shouldn't leak current to chassis which is a defect in itself.

You could make a short cable with a wall type receptacle on one side and a wall type plug on the other side. Make it out of 3 wires, where you connect L to L, N to N and earth to earth, attach an ampere meter in series on each of the wires.
Then you can go around your house with all appliances disconnected and connect them to your ampmeter cable one at a time.

If truly one of your appliances is faulty itself then you should see a discrepancy between the amp readings of that appliances live and neutral.

This is the same thing a ground fault circuit interrupter does, it compares the current between live and neutral because if everything is ok those currents should exactly match, if they don't there can be only one reason and that is the device is leaking current to ground.

 

[Rive]

Erm. I think that by throwing in DIY equipment for investigating a live line voltage electric network in unclear state we are treading on thin ice here.

 

[Averagesupernova]

Erm. I think that by throwing in DIY equipment for investigating a live line voltage electric network in unclear state we are treading on thin ice here.

I think you might be surprised that the "DIY equipment" AKA the incandescent bulb is a very common tool amongst electricians. Do you have another tool in mind?

 

[Rive]

I know that and I think I would be able to dig up mine from some boxes in storage: I'm myself also fine with extension cords with patched in current measurement (though I prefer clamps instead) but we are in the open here.

 

[Wrichik Basu]

I am going to ignore this thread until the filament lamp test has been done.

Connect a filament lamp between Earth and Neutral.
Does the lamp glow ?
What is voltage across the lamp, before and after ?

Once you have performed that test, it will be possible to work out the next step in the search for what must be at least two faults.

The first test I would do is to connect a 240V filament lamp between earth and neutral, then measure the voltage again.
1. If the breaker immediately cuts that circuit, then you have a good RCD breaker.
2. If the voltage then falls toward zero, you are missing your local ground to neutral link. Call an electrician to restore that link in your distribution box.
3. If the lamp glows red then wires are crossed in your vicinity, you have a problem that certainly needs an electrician to resolve the safety issue.

I have a 15 W incandescent bulb. Connected it b/w earth and neutral. It didn't glow. (The 0.5 W LED bulb glows, however.)

PD before connecting lamp: 29.84 V
PD after connecting lamp: 30.12 V.

If you want me to get a bulb of lower rating, I will have to search for it.

 

[Wrichik Basu]

Erm. I think that by throwing in DIY equipment for investigating a live line voltage electric network in unclear state we are treading on thin ice here.

I know that and I think I would be able to dig up mine from some boxes in storage: I'm myself also fine with extension cords with patched in current measurement (though I prefer clamps instead) but we are in the open here.

Please don't report the thread for high voltage safety issue. My multimeter probes are rated for 1000 V, 10 A max. I use proper safety precautions when measuring mains voltage.

 

[Averagesupernova]

I have a 15 W incandescent bulb. Connected it b/w earth and neutral. It didn't glow. (The 0.5 W LED bulb glows, however.)

PD before connecting lamp: 29.84 V
PD after connecting lamp: 30.12 V.

If you want me to get a bulb of lower rating, I will have to search for it.

Assuming the 15 watts lamp is not defective even though it does not glow indicates there is a significant leak someplace that is partially energizing the earth conductor since you measured voltage.
-
You didn't previously mention that the 1.5 watt was an LED bulb. These are not good for this type of troubleshooting.

 

[Wrichik Basu]

You may know that each earth pin is supposed to connect back to earth but do you really know the route? I would be familiar with how things are supposed to be done in the USA but what is the practice where you are at? In the USA the earth conductor is run in the same cable or raceway as the hot and neutral back to the source. If there is a broken earth wire in the raceway only devices downstream from that break will be affected. It's not spaghetti when done correctly. Based on your diagrams, I have to wonder about your system.

The earth wires from the rooms are joined to each other at the nearest junction boxes inside the house. These wires then finally meet at the main junction box, from where the wires descend downstairs. As I had said in post #11,

It seems that the earthing wire(s) that go down from our house to the circuit board downstairs are somehow damaged or have got disconnected.

This is what the electrician will locate, if he concurs with my inferences.

---

You didn't say the 1.5 watt was an LED bulb. These are not good for this type of troubleshooting.

Please see post #11:

Tried that with a 0.5 W LED bulb.

 

[anorlunda]

In the USA, our answer would certainly be to leave this problem to a licensed electrician; trusting the electrician to follow codes. In India, that may be less practical.

However, I'm uncomfortable about recommending some of these DIY troubleshooting methods. I fear safety hazards. @Wrichik Basu is a valued member of PF. I would not want to see him hurt.

Is the advice given prudent?

 

[Averagesupernova]

In the USA, our answer would certainly be to leave this problem to a licensed electrician; trusting the electrician to follow codes. In India, that may be less practical.

However, I'm uncomfortable about recommending some of these DIY troubleshooting methods. I fear safety hazards. @Wrichik Basu is a valued member of PF. I would not want to see him hurt.

Is the advice given prudent?

Several things. @Wrichik Basu has been given sound advice here. Closing this thread will not likely stop him from continuing to investigate this. An electrician called in could potentially end with the electrician saying to replace the fridge, everything else is fine. You said yourself that in India getting an electrician may not be practical. That implies instead of @Wrichik Basu investigating this, someone less qualified does it.

 

[Wrichik Basu]

Well @Wrichik Basu , no offense but the pictures you provided show a rather dirty fuse box and the neutral and earth connections do seem oxidized and rusted outright, that can be a problem or part of it, but to know one would need to measure the contact resistance and visually inspect them thoroughly.

For the neutral, the company needs to be called. And they are least interested in these things. For the earth, I need to have a word with my neighbours. And I would shoot myself before trying to make them understand the issue.

You could make a short cable with a wall type receptacle on one side and a wall type plug on the other side. Make it out of 3 wires, where you connect L to L, N to N and earth to earth, attach an ampere meter in series on each of the wires.
Then you can go around your house with all appliances disconnected and connect them to your ampmeter cable one at a time.

While a good way to find the defective device, I am a bit hesitant to follow this at this moment for the following reasons:

1. My multimeter has a hall clamp (for both AC and DC), so I don't need to connect the multimeter in series. However, it's a lot of work, to be honest.
2. I want to "preserve" the fault till the electrician comes in. If I rectify it, then I will not be able to point out that the earth is faulty.
3. From my data in post #27, I believe we can rule out the refrigerator as a culprit, since there is not a lot of difference between the PD measured when it is plugged in vs. unplugged. (The fridge is also brand new, bought just a few months back…) All other appliances we own have a plastic body, so there is no chance of shock to the user (except that one immersion heater, which we unplug after use).
4. Once the earthing is done properly, if some appliance breaks down by drawing excessive current into the earth and blows a fuse, let it do so. The house is well protected by properly rated fuses at different points. And it will be easier to locate it at that time.

---

In the USA, our answer would certainly be to leave this problem to a licensed electrician; trusting the electrician to follow codes. In India, that may be less practical.

However, I'm uncomfortable about recommending some of these DIY troubleshooting methods. I fear safety hazards. @Wrichik Basu is a valued member of PF. I would not want to see him hurt.

Is the advice given prudent?

I understand and appreciate your concerns. But in India, unfortunately, there is no such thing as a licence for being an electrician, let alone codes to follow. I really wanted to document the unsafe ways they adopt in their everyday work. Following advice from experienced users in PF is a lot better than letting them tear down each and every socket, trying to find a fault. I believe doing measurements with a multimeter (with safety measures adopted) and using a bulb won't cause a lot of harm. The rest is up to the mentors to decide.

I will be calling an electrician in the next few days, and hopefully, he will be able to solve the issue. Learnt a lot through this thread. Thanks, everyone, for your time and input!

 

[Averagesupernova]

From my data in post #27, I believe we can rule out the refrigerator as a culprit, since there is not a lot of difference between the PD measured when it is plugged in vs. unplugged. (The fridge is also brand new, bought just a few months back…) All other appliances we own have a plastic body, so there is no chance of shock to the user (except that one immersion heater, which we unplug after use).

I disagree here. If the appliance puts voltage on the earth conductor it should be considered defective. However, you may measure voltage that is capacitively coupled onto the earth conductor if it is not properly connected to earth. The same thing could happen if you toss a length of wire on the floor and measure the voltage on it relative to earth. BUT, that does not mean it can source any current. If it cannot source current, that means it won't shock you. Since you are getting shocked, something somewhere is defective. Having the bulb connected when doing the measurements indicated a voltage which means current can be sourced.

 

[Rive]

Please don't report the thread for high voltage safety issue.

I won't, but since we are in the open too many DIY ideas might leave no room for consideration for mentors.

If the appliance puts voltage on the earth conductor it should be considered defective.

Well, actually...

Many electronic appliances starting with an EMI filter like this:

With a broken PE connection this kind of filter will provide ~ half line voltage on the PE pin (and, consequently: on the chassis). Most common case is a PC hanging on a cheap, inconsiderately chosen extension cord, giving pricking sensation for any light touch.

With several appliances (on different phases) hanging on a common (broken) PE the voltage measured on the PE may be ... interesting.

I would prefer having a voltage measured between PE and water tap (if there are metal pipes for water there). With and without a (filament) bulb connected between them.

 

[Averagesupernova]

I would prefer having a voltage measured between PE and water tap (if there are metal pipes for water there). With and without a (filament) bulb connected between them.

Let's not confuse this any farther. Measure the voltage between the points we know are not supposed to have voltage between them. I don't know if anyone here knows if plumbing is connected in any way in India or not. At least I don't.

 

[Rive]

I don't know if anyone here knows if plumbing is connected in any way in India or not.

Old style (metal) water pipes usually works as reference. And with the official PE-ground connection in that shape (and without proper equipment) I see no other way to get an indication (not measurement! just indication) for the quality of the actual PE<>ground relations.

Ps.: honestly, based on the pictures, I feel like the pipes might be the most reliable stuff there

 

[DaveE]

I avoid posting in threads like these because I always question the accuracy of the system as it is described.

I am describing what the company had described to me the last time they came for an inspection (many years back, but nothing has changed since then).

Yes, but how do you really know that either they were correct or that nothing has changed? You (someone) needs to verify what's really there first.

I was an R&D EE at a company that made some complex high power electronic instruments. Occasionally I'd be asked to help troubleshoot systems on the manufacturing floor. They always fixed the simple stuff themselves, of course. I always started by asking them what was happening, what did they think, etc. Then I mostly ignored their observations and remeasured everything from the beginning. Almost always, they were assuming something that wasn't actually true, even though it should have been. Often the problem was simple once you got past the otherwise reasonable assumptions.

PS: sorry, I know that's what you are doing. But keep in mind the problem isn't so much the fix, the problem is understanding what to fix. It is so easy to lose sight of that. It was a frequent problem for me in the lab troubleshooting my own designs too.

 

[Averagesupernova]

Almost always, they were assuming something that wasn't actually true, even though it should have been.

Ding ding ding ding ding!
-
This is why we start troubleshooting at the place where measurements and function is working correctly and makes sense.

 

[Baluncore]

When hot, the current through a 15 watt, 240 volt filament will be about 15 / 240 = 0.0625 amp. When cold, the resistance will be one tenth, so the current will be 0.625 amp. The cold filament resistance will then be about 240 / 0.625 = 384 ohms.

When Earth and Neutral are connected by the cold filament of 384 ohms, the voltage RISES from 29.84 V to 30.12 V. That is an increase of 0.28 volt.

I would immediately check the series connections in the Neutral supply.
How much of the E-N fault voltage appears across the Neutral fuse, and its poor connections?

 

[Averagesupernova]

When Earth and Neutral are connected by the cold filament of 384 ohms, the voltage RISES from 29.84 V to 30.12 V. That is an increase of 0.28 volt.

Hmmm. I at first read this the opposite. But that does indicate a weak neutral. Unless @Wrichik Basu accidentally reported the readings to us reversed. I'd like a verification there.

 

[Baluncore]

I'd like a verification there.

If it was a half volt drop, would it change the analysis?

 

[Averagesupernova]

If it was a half volt drop, would it change the analysis?

Only if loaded/unloaded is repeatable.

 

[OmCheeto]

... after some appliances were giving electric shocks...

Other than the immersion heater, can you list the appliances giving shocks.
And do they shock you if the immersion heater is unplugged?

 

[Wrichik Basu]

When Earth and Neutral are connected by the cold filament of 384 ohms, the voltage RISES from 29.84 V to 30.12 V. That is an increase of 0.28 volt.

Hmmm. I at first read this the opposite. But that does indicate a weak neutral. Unless @Wrichik Basu accidentally reported the readings to us reversed. I'd like a verification there.

The readings were fluctuating somewhat. But is 0.28 V really something you should be worried about? It may be a measurement error.

Other than the immersion heater, can you list the appliances giving shocks.
And do they shock you if the immersion heater is unplugged?

My laptop is giving a tingling sensation. The metal parts of the refrigerator body are sometimes giving a tingling sensation. All other appliances are basically plastic or fibre glass body. I started investigating from the bathroom outlet, but later found that all outlets have this PD.

The immersion heater is giving shock independent of whether it is switched on or not. It simply has to be plugged in to the outlet, and you will start getting the shock.

---

My question, at this juncture, is, is it normal to have a > 28 V PD b/w earth and neutral at all times? When live-neutral measures 236 V, live-earth measures 182 V. Is this even normal?

You know what? I wouldn't be surprised if the electrician, at the end of the day, says, "That's very normal, I have seen it in other houses too."

 

[Baluncore]

My question, at this juncture, is, is it normal to have a > 28 V PD b/w earth and neutral at all times? When live-neutral measures 236 V, live-earth measures 182 V. Is this even normal?

It is not normal. It is dangerous.

Too much talk and not enough tests.
Measure the voltage drop across the fuse in the neutral line.

Somewhere nearby, the neutral will be earthed at a pole or a distribution transformer. Look for a bare copper wire running down a power pole. Can you find that place, or are the lines so unregulated that you cannot tell what has been done?

 

[Wrichik Basu]

Measure the voltage drop across the fuse in the neutral line.

It's negligible, $\approx 0.003~\mathrm{VAC}.$

Somewhere nearby, the neutral will be earthed at a pole or a distribution transformer. Look for a bare copper wire running down a power pole. Can you find that place, or are the lines so unregulated that you cannot tell what has been done?

The wires pass under the ground, and they come up only at the local transformer and the houses. Other than, they are not visible anywhere, so I cannot visually see the link between the neutral and earth lines.

The PD between the neutral fuse (point D) and the earth lines (point E) is around $1.27~\mathrm{VAC}$ (measured it today again).

Also wanted to mention one fact: if I connect the 15 W incandescent bulb b/w live and earth in our house, it does not glow at all. It's not faulty because it glows fine when connected normally between live and neutral, but does not glow when connected b/w live and earth or earth and neutral.

The 0.5 W LED bulb, however, glows when connected b/w live and earth, but with diminished brightness.

 

[Wrichik Basu]

Here are some more pictures of the region near our main fuse box (F1). Thought it might help illustrate the region more.

The wires as they enter our fuse box (F1).


The innards of the fuse box.
​

The PD between the joint on the side of the fuse box (where our green earthing wire joins with the second red wire) and the neutral is $\approx 1.2~\mathrm{VAC}.$

 

[Wrichik Basu]

Yes, but how do you really know that either they were correct or that nothing has changed? You (someone) needs to verify what's really there first.

The diagram I posted in the OP should be correct because, as I had noted, there are the electricity meters between the company fuse (A-C) and the fuses for each house (F1-6). I had omitted these meters for the sake of simplicity. Changing anything below and including the meter is illegal, and that is managed by the company only. We, as citizens, have the right to change things above the meter, i.e. from F1 and above that.

 

[Rive]

Also wanted to mention one fact: if I connect the 15 W incandescent bulb b/w live and earth in our house, it does not glow at all. It's not faulty because it glows fine when connected normally between live and neutral, but does not glow when connected b/w live and earth or earth and neutral.

By earth, you mean the PE line, right?

At this point it is just an assumption that PE is (still) properly connected.
Could you please make the measurement I requested from #43?
Be sure that during the measurement your computer is connected and working.

 

[Wrichik Basu]

By earth, you mean the PE line, right?

What does PE mean? (Sorry, not accustomed to this acronym.) I meant the earth in the 3-pin outlet.

 

[Rive]

What does PE mean? (Sorry, not accustomed to this acronym.) I meant the earth in the 3-pin outlet.

That line is called 'Protective Earth' (PE for short). From the story so far I'm assuming that it's supposed to be connected to Earth at a specific point (and: to Neutral too), but that's only if the system is up to Code.

Since the system is clearly not up to any Code, to call it 'Earth' requires confirmation first.