Investigating a faulty earthing in our house

[Wrichik Basu]

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.

OK, so I took data at two different points, one in the kitchen and one in the washroom.

In the kitchen, the PD between the earth pin in the 3-pin outlet and the water tap was $\approx 2.1~\mathrm{VAC}$ while in the washroom, it was $\approx 1.3~\mathrm{VAC}.$ Note that these taps aren't directly touching the ground; they come down from the overhead reservoir.

For record, the neutral-earth PD at the time of measurement was $\approx 14~\mathrm{VAC}.$ Yes, previously I had been stating the value to be 28 VAC but it seems that the value deceases at night and again increases in the morning. I have noted this for two consecutive days. Edit: I was wrong. It later rose to $31~\mathrm{VAC}.$

Also, does the fact that the incandescent bulb doesn't glow when connected b/w live and earth have any significance?

 

[Rive]

Note that these taps aren't directly touching the ground; they come down from the overhead reservoir.

Then this measurement was a dead end. Sorry.

The point would be to check the voltage on the PE line against some 'real' ground: against some metal pipes what comes up directly from ... ground.

If the PE line is properly connected, then no voltage is expected. If it is 'floating' then there will be some voltage due those EMI filters I mentioned somewhere before: but those EMI filters won't allow much current to flow, so the voltage would disappear when a bulb is connected.

For this measurement if the bulb lights up then the the whole network is dangerous and you should run and remove the fuses immediately, because not only the PE floating, but some faulty appliance actually feeding current to it - making the whole 'protection' of the network deadly.

Also, does the fact that the incandescent bulb doesn't glow when connected b/w live and earth have any significance?

That experiment might also indicate a floating PE, but when executed with an actually floating PE, this experiment may create a life threatening danger through bringing PE (and all connected device chassis across the network) to line voltage.

 

[Wrichik Basu]

The point would be to check the voltage on the PE line against some 'real' ground: against some metal pipes what comes up directly from ... ground.

Unfortunately, I haven't got any such pipes inside my house. Such pipes are available in the basement, near the main fuse board. But there, I have already checked that the PD b/w neutral and earth connections is $\le 2~\mathrm{VAC}.$

For this measurement if the bulb lights up then the the whole network is dangerous and you should run and remove the fuses immediately, because not only the PE floating, but some faulty appliance actually feeding current to it - making the whole 'protection' of the network deadly.

As I said, the $15~\mathrm{W}$ incandescent bulb doesn't light up, but the $0.5~\mathrm{W}$ LED bulb does. Probably if I can get hold of an incandescent bulb which is of very low rating, then that will light up.

 

[OmCheeto]

...

Also, does the fact that the incandescent bulb doesn't glow when connected b/w live and earth have any significance?

I would say definitely yes.

My laptop is giving a tingling sensation.
...

This is interesting, as I experience the same thing when I simultaneously operate my two metal cased MacBook laptops. One of them is less than a year old, and I seriously doubt there is an electrical problem with the other one. I may get out my O-Scope and see what's up with that.

And although I can't recreate it with my appliances, I remember that my mother's 1960's era stove and refrigerator gave the most entertaining 'tingle' when you touched them both at once, but not individually.

How old is your refrigerator?

 

[Wrichik Basu]

How old is your refrigerator?

Brand new, bought a few months back. It doesn't always give the tingle as most of its body is made of non-conductive material, but sporadically, it does if I touch its back.

 

[Rive]

I have already checked that the PD b/w neutral and earth connections ...

Please specify: did you check neutral to PE or neutral to ground?
Also, how sure you are that the PE in the basement is the same (=well connected) than the PE you have up?

 

[Wrichik Basu]

Please specify: did you check neutral to PE or neutral to ground?

Neutral to PE. If you refer to the diagram in the OP, point D to E.

Also, how sure you are that the PE in the basement is the same than the PE you have up?

Exactly. The PE wires in the basement give a potential difference of $\le 2~\mathrm{VAC}$ when measured against neutral (which is acceptable), but the ones inside our house have a different and much higher value. I believe something is messed up/snapped in the wiring between the basement and our house, and that is what the electrician has to find.

 

[hutchphd]

but does not glow when connected b/w live and earth

The neutral should be hard wired to the earth circuit somewhere. Enough to light a 15 W bulb. Or do I not understand British electricity? Comments please.

 

[OmCheeto]

The neutral should be hard wired to the earth circuit somewhere. Enough to light a 15 W bulb. Or do I not understand British electricity? Comments please.

According to wiki, Indian "Earthing is to be done with two separate connections."
My house here in the states has two separate connections.

Wrichik, what is your plumbing made of? Mine is a mixture of iron and plastic, with a copper jumper installed where a plastic piece fixed a broken '2nd earth section' iron pipe, many decades ago.

 

[Wrichik Basu]

According to wiki, Indian "Earthing is to be done with two separate connections.

That's something interesting you found. So we do have some guidelines. That will be useful in the future.

Wrichik, what is your plumbing made of?

Mainly metal pipes, but some joints have been replaced with plastic counterparts.

 

[OmCheeto]

In our house, the potential difference (PD) between neutral and earth is 50−60 VAC, which should normally be <2 VAC if earth is properly connected.

Going back to your original post, I'd like to point out that "should normally be <2 VAC" is not necessarily true. I did a search of this thread and the keywords 'ghost' and 'phantom' are missing, so I'm guessing no one has mentioned this yet. Rather than blather an explanation, I'll simply quote a source I found;

"Due to the high impedance of measuring instruments, a voltage reading may be detected on open conductors where there is no hard electrical connection to a voltage source. Conductors that are installed in close proximity to one another, and are capacitively coupled to each other, can cause this a.c. voltage reading. Such a reading could be 2 or 3 volts, or it may be as high as the voltage on the adjacent conductors. This is what is referred to as a “phantom” voltage." (ref)​

bolding mine​

In other words, your 50-60 VAC is perfectly normal, as long as it's a phantom or ghost voltage.
This is of course what some have been prompting you to check for, without actually calling it by it's names.

This is not to say nothing is wrong. I'm pretty sure the ground wires in your house are not actually grounded. Or at least the one you tested with the incandescent lamp isn't.

 

[Baluncore]

This is of course what some have been prompting you to check for, without actually calling it by it's names.

Yes.
That is why we use a filament lamp as a load when measuring a voltage. A filament lamp will not be damaged if you connect it across the full supply. Using a lamp when troubleshooting often eliminates the need to measure voltage. An LED lamp is NOT a substitute for a metal filament.

 

[Averagesupernova]

I did a search of this thread and the keywords 'ghost' and 'phantom' are missing, so I'm guessing no one has mentioned this yet.

See post #42.
Edit:.

This is of course what some have been prompting you to check for, without actually calling it by it's names.

Missed that part when I first read your post. My apologies.

 

[Averagesupernova]

Let's leave the water pipe out of this. We know there is little to no voltage between the neutral and earth close to the source. Voltage develops for some reason in the house. Only cause can be an open or partially open conductor. If the house had absolutely no plumbing the same problem could develope and it would still have to be troubleshot. The solution would not be to install plumbing and then take measurements.

 

[Wrichik Basu]

Going back to your original post, I'd like to point out that "should normally be <2 VAC" is not necessarily true. I did a search of this thread and the keywords 'ghost' and 'phantom' are missing, so I'm guessing no one has mentioned this yet. Rather than blather an explanation, I'll simply quote a source I found;

"Due to the high impedance of measuring instruments, a voltage reading may be detected on open conductors where there is no hard electrical connection to a voltage source. Conductors that are installed in close proximity to one another, and are capacitively coupled to each other, can cause this a.c. voltage reading. Such a reading could be 2 or 3 volts, or it may be as high as the voltage on the adjacent conductors. This is what is referred to as a “phantom” voltage." (ref)​

bolding mine​

In other words, your 50-60 VAC is perfectly normal, as long as it's a phantom or ghost voltage.

If I consider, for the sake of the argument, that the potential appearing on the PE is a phantom voltage, I guess that establishes that the PE is actually not grounded, because the phantom voltage can only occur in open circuits, as per your source.

 

[Wrichik Basu]

Yes.
That is why we use a filament lamp as a load when measuring a voltage. A filament lamp will not be damaged if you connect it across the full supply. Using a lamp when troubleshooting often eliminates the need to measure voltage. An LED lamp is NOT a substitute for a metal filament.

OK, so I took readings across the 15 W filament lamp one more time, just to confirm the readings. Note that the bulb glows normally when connected between live and neutral.

Connected the lamp across neutral and PE.

PD immediately before connecting: $29.58~\mathrm{V}.$
PD when connected:$0.108~\mathrm{V}.$ (The lamp didn't glow.)
PD immediately after disconnecting: $29.56~\mathrm{V}.$

 

[Baluncore]

If I consider, for the sake of the argument, that the potential appearing on the PE is a phantom voltage, I guess that establishes that the PE is actually not grounded, because the phantom voltage can only occur in open circuits, as per your source.

If the PE was open circuit, then the PE-N voltage would have fallen significantly when the filament lamp was connected across the meter.
So I think it is neutral current through resistance that is causing part of the problem.

If you could run a wire back to the missing PE-N junction, you could measure the voltage at your house, (with filament globe in parallel) and so determine if it was N or PE that was resistive.

Instead, you might connect an electric water heater to the power outlet and see how much drop you get in the mains voltage. That will allow you to compute the resistance of the A-N circuit return.

 

[Baluncore]

PD when connected: 0.108 V (The lamp didn't glow.)

The filament lamp exorcised the ghost voltage.
Your PE is open circuit. Get it fixed for your own protection.

 

[Guineafowl]

OP’s house appears to have TT (Fr. terre-terre) earthing, specifically pipe earthing. There is no N-E link except at the company transformer.

It’s common for N to float a volt or two above E, if current is being drawn. This is the case at the service head, but at the OP’s house, the voltage has risen to dangerous levels.

This rogue N-E voltage sags when loaded with a bulb, but is enough to cause tingling shocks. This suggests a poor connection somewhere in the E wire from house to service head. There may also be some earth leakage from a number of faulty appliances.

Hopefully, the electrician will trace the bad connection, or simply run a new E wire to OP’s house.
Edit: Ideally, he should check earth leakage (Megger L+N to E, or clamp L+N while running) on suspect appliances too.

 

[Wrichik Basu]

An electrician will be visiting today. Let's see how things turn out.

 

[Wrichik Basu]

Final update

The electricians confirmed our hypothesis that the PE line was not grounded. They opened up all the junction boxes, from our house to the main fuse board downstairs, and found that extremely low quality, thin and single core wire had been used for earthing. This same quality wire was used for earthing for both the low current outlets, as well as the high current $(11~\mathrm{A})$ outlet. A few pictures of the wire when they took it out:

​

Note that it was already snapped when they removed it. Also notice the condition of the insulation.

So, we went to our local trusted electric supply store and bought good quality $1.5~\mathrm{mm}^2$ wire, as recommended by the electricians. A comparison of the old and new wires can be seen below:

The previous wire was completely replaced, from our house to the main fuse board, with the new wire. Thereafter, I checked the PD between neutral and PE with my multimeter; it was around $0.3~\mathrm{V}$ in all outlets.

The electricians also showed that the earthing wires inside our house are made of that same low-quality wire. We will be changing them in the coming days.

Once again, PF has helped me a lot in diagnosing the problem. Also learnt a lot in the process. Thanks to everyone who participated in helping me out, especially @Baluncore, @Averagesupernova, @Lnewqban, @Rive, @hutchphd, and @OmCheeto for your precious time and input. Also, a massive salute to the moderation team, without whom the site would not have been functioning as it is today.

 

[Rive]

I've seen some terrible wirings already, but that's more or less the worst so far

Ps.: PE corroding away - happens. PE not present - there was such wiring standard here 'once upon a time'. Everything with Al wires, with slowly degrading connections - well, it was common once.
But this is ~ about intentionally providing false security which seems to work but melts away by the first trouble - evil

 

[Wrichik Basu]

I've seen some terrible wirings already, but that's more or less the worst so far
...
But this is ~ about intentionally providing false security which seems to work but melts away by the first trouble - evil

(Abridged by me)[/size]
Yeah, it is kind-of evil, frankly. While changing the wires, we found that only our house had that kind of low quality wires for earthing. There are possible reasons for that, but I'll not make the thread off-topic, except stating that the builder had opted for a nasty and criminal way to save money.

 

[hutchphd]

I've seen some terrible wirings already, but that's more or less the worst so far

I was fixing up a house in Maine which had been built in the early 50's. I found some newer Aluminum wiring at an outlet box and discovered that some fool had twisted solid Al onto 12 gauge copper Romex, covered it with "friction tape" and left that connection unsheathed in the wall cavity. There was no corrosion at the junction (ergo the house was still standing!), but I looked at every wire in the house immediately thereafter. caveat emptor