Where's the Earth ?

[michael8554]

Intending to have the house CU upgraded in the New Year, for safety reasons and to allow future expansion - EVSE etc.
For my own satisfaction I have labelled the wired fuses and worked out what goes where, but haven't been able to find a house earth.
It's a 1930's detached house on the outskirts of a country village, "rewired" in the 1970's.
On the image:



A Two incoming overhead cables
B Two 100Amp fuses - no earth connection ??
C Off-Peak Timer and Contactor
G 100Amp 30mA RCD, two live outputs, to Storage Heaters and House CU
D Live and Neutral to house CU, and a grey T&E from H, the PV Inverter Isolator and meter.
E, F Thin green and green/yellow cables

In the kitchen there is a thin bonding cable attached to the incoming water main, it disappears to who knows where, but not to the CU as far as I can see, there isn't a gas main.

So no obvious thick main earth attached.

Do you think there is an earth spike somewhere ?

Does the RCD cover the apparent lack of an earth ?

The PV electrician made no mention of the earthing.

Thanks for looking

Michael

 

[Luckyboy]

Are you able to see where "F" wanders off to?
Assuming this is in your garage?

 

[Andy™]

looks like a TT supply. there should be an earth electrode, usually a rod somewhere. possible at that age there isnt one and its relying on the water / gas

the electrician who does the work will be able to verify what is / isnt there and what needs to be done

 

[ProDave]

Is this in the UK?

The earth is probably TT, probably a bit of 6mm connected to an exposed terminal on an exposed rod, next to the back door that has long since corroded away.

I would be even more concerned with wanting to understand, and probably upgrade an old and complex metering system. Why 2 supplies? Never seen a separate supply into the property for the off peak. Is it a mansion?

It looks like a dual meter with 2 dials, not a type of meter I have seen (which is why I ask is it UK) It could be a version of the very old white meter system that pre dates Economy 7 (and would tie with a 1970;s date)

 

[michael8554]

Thanks for the speedy replies.

"F" comes from the house CU, and goes to the round brown bakelite terminal block to the left of "F" in the DIY-looking garage lighting circuit.

Yes, all in the attached garage.

Yes UK.

Note the two 6-way Wylex fuse boxes for the night storage heaters, of which 8 ways are occupied, is that a significant load on it's own ?

But way back, the large "back garden" was a pig farm, so maybe it was originally 100A for the house, 100A for the farm equipment ?

If there was an earth spike, shouldn't it be attached to somewhere in my image with a 6mm2 or bigger cable ?

Could be coming through the brickwork into the back of the house CU, I can see a cable bundle behind the CU board.

Was the 100A RCD added to cover the earthing ?

 

[binky]

tidy installtion, old wooden wylex boards, looks ike it hasn't been touched since being rewired in the 70s, apart from a new main fuse carrier, and addition of the 100A RCD. As said above, earthing is probably via water supply pipe, or 'lost' earth rod of some sort. I suspect you may find your ligting has no earhting aswell.

 

[michael8554]

To sort of answer my own question, I measured AC Volts between Live and Neutral terminals, and between Live and Earth terminals, on a selection of the house sockets.

Live to Neutral was 1 volt lower.

Not as good a test as a Fluke Tester would do, but reassuring - or is it ?

 

[Andy™]

measuring voltage L-N, L-E etc is completly useless and poves nothing about the earthing

 

[michael8554]

Thanks Andy.

Does it prove there is a low resistance earth, which is not necessarily beefy enough to take a big short current ?

Michael

 

[Andy™]

no. you need something that can measure impedance, not voltage. completely different

 

[michael8554]

Thanks again Andy

My reply did not compute :-<

A length of T&E, with L and N connected to mains, but earth not connected to any sort of earth eg water main or earth spike etc,.

Would I be able to measure 244V between L and E ?

Does this indicate there is some sort of earth connection, but its impedance isn't known ?

 

[apprentice87]

Thanks again Andy

My reply did not compute :-<

A length of T&E, with L and N connected to mains, but earth not connected to any sort of earth eg water main or earth spike etc,.

Would I be able to measure 244V between L and E ?

Does this indicate there is some sort of earth connection, but its impedance isn't known ?

The voltage means nothing,, Have a study of Ohms law.. You could have a reading of 244V [or whatever] between L and E but think what would happen if the RESISTANCE [yes yes, we all know] of said earth was 50,000 ohms.. No, make that a MILLION ohms.. You would still have a reading of 244V, but say you had a fault where the casing of an appliance ended up touching a live wire?? Current would then flow to earth, yes?? BUT how MUCH current.. Well, it would be 244V/1,000,000 ohms so 0.000244 amps.. Now, can you see that this amount of current woulf be nearly EIGHT THOUSAND time TOO LOW to trip the RCD never mind blow the fuse.... Do you see now?? The earth has to have a resistance [yes, i know for the pedants out there] low enough to flow enough current to blow the fuse or trip the RCD... If it does not, the faulty appliance will remain at full mains voltage for ever [which will render you very dead!!]

Hope this helps...

 

[GEM Electrics]

A very interesting setup, and after an hour studying the photo I still cannot make out all the connections.

The first point is that under modern regs the earth cable for a TT supply, assuming protection against corrosion, need only be 4mm if unprotected (2.5mm if machanically protected) so if it were a modern TT system you would not be looking for a 6mm or more earth. But is it a modern TT system? Back in Ed 5 days (which looks typical of this installation) 7/044 earth would have been quite acceptable. Also in those days plastic water pipe was unheard of and using the incomming lead/copper water pipe for the earth was a really good idea. So the "bonding cable attached to the incoming water main" may well not be bonding but the main earth coming into the original Fuse Box at "F". However I would have expected it to be connected to the two Night Storage Fuse Boxes as well. (Yes Fuse Box not Consumer Unit).

"Does the RCD cover the apparent lack of an earth ?" well to some extent, since in the TT system only small earths are required as above.

Now why two supply cables? Well those Night Storage units could easily be 2.5KW, over 10Amps each, or 83Amps for 8 so yes they needed their own supply. I have seen an installation where the total load was taken via a single phase supply cable where the insulation of the cable at 11.30pm literately melted and had to be upgraded to a 3phase supply. Which leads me on to think that this may be two single phase supplies but different phases, and that leads on to suspect that the meter is a three phase meter and the RCD (at G) is a 3 phase RCD. Note that the RCD has two inputs as well as 2 outputs. EXCEPT THAT the two Night Storage Fuse Boxes seem to be feed from different supplies. However the overloading principle still applies.

NB If this is so then there are 440Volts floating around so this should be checked and labeled.

I have not yet seen any comment on the colour of the cable in the righ hand phase block which appears to be brown, however if you trace this cable back it is the Grey cable going to the right hand Night Storage Fuse Box. So I would suspect that the colour indicates that this (neutral) cable is seriously damaged by over heating. Note that the corresponding live cable, comming from contactor box, is also changing color. Are these cables somewhat overloaded?

If the Fuse Box(es) are replaced with modern Consumer Units then the system as a whole will need to comply with Ed18 and that will require:

• A new earth for the whole system (Earthing through the water pipe is no longer allowed)
• Bonding of the water if the existing "earth" is inadequate
• Changing the 3phase 30mA RCD for an S type (100mA) 3 phase (very expensive?)
• Replacing the damaged cable
• Using 30mA RCBO's on all circuits in the new CUs (Reg 314.2)

(What have I forgotten?)

Of course, circuits that are not modified do not need to be upgraded so long as the installation is in accord with the regs that applied at the time of installation (Ed5?) but upgrading a Fuse Box to a CU is classed as a modification.

 

[GEM Electrics]

The voltage means nothing,, Have a study of Ohms law.. You could have a reading of 244V [or whatever] between L and E but think what would happen if the RESISTANCE [yes yes, we all know] of said earth was 50,000 ohms.. No, make that a MILLION ohms.. You would still have a reading of 244V, but say you had a fault where the casing of an appliance ended up touching a live wire?? Current would then flow to earth, yes?? BUT how MUCH current.. Well, it would be 244V/1,000,000 ohms so 0.000244 amps.. Now, can you see that this amount of current woulf be nearly EIGHT THOUSAND time TOO LOW to trip the RCD never mind blow the fuse.... Do you see now?? The earth has to have a resistance [yes, i know for the pedants out there] low enough to flow enough current to blow the fuse or trip the RCD... If it does not, the faulty appliance will remain at full mains voltage for ever [which will render you very dead!!]

Hope this helps...

Not to detract from the point that you are really making (that the earth needs to be an earth and not just a floating wire). I am told that the reason we have 32mA RCDs or RCBOs is that this is the limit of current that a person can withstand for a period of 40mS that would cause a fatality. I don't know how true that is but a current of 0.000244 amps, while very unplesent, is unlikely to be fatal. cf the current passing from an electric fence at maybe 10,000 volts.

 

[apprentice87]

Not to detract from the point that you are really making (that the earth needs to be an earth and not just a floating wire). I am told that the reason we have 32mA RCDs or RCBOs is that this is the limit of current that a person can withstand for a period of 40mS that would cause a fatality. I don't know how true that is but a current of 0.000244 amps, while very unplesent, is unlikely to be fatal. cf the current passing from an electric fence at maybe 10,000 volts.

You have missed the point entirely, re-read my post..

If the EFLI is 0.000244 amps, what will happen in the event of an earth fault?? [or more to the point not happen] Now, if a person touches an exposed conductive part in these circumstances, it does NOT mean that the current that flows through THEM will be 0.000244 Amps. It will be more like 0.25 of an amp... What will happen to them??

john..

 

[michael8554]

Thanks everyone who replied, I'm really enjoying this :->

From Andy's post:
I understand Ohm's Law, and that a MILLION OHMS in the E wire would not be enough to trip the 30mA RCD.
And that the person touching the Live case of a faulty appliance would not receive only 244uA, they would get >30mA and so get a jolt.
I get that the potential on L is always 244V, but if a DVM has one probe on L and the other touching nothing, it wouldn't read 244 Volts.
So a reading when the probe touches E indicates a measurable resistance, although in all likelihood that could be dangerous Meg Ohms ?

GEM Electrics.
Thanks for the comprehensive examination. Here's the Meter:



It's a Landis and Gyr ZMB 127 3-Phase meter.
The leds above "ZMB127" and "1992" are both on and amber.

The RCD is a 4-Pole Wylex WRS 100.

I take your point that the grey N is brown nearest the RH Henley - a loose connection in there?
The RH Night Store CU has all 6 fuses in use so draws the most amps.
But in use the cable doesn't feel warm to the touch.

 

[Andy™]

NB If this is so then there are 440Volts floating around so this should be checked and labeled.

label is required if voltage to earth exceeds 230v, voltage between phases is irrelevant. not something your likely to find, especially in a house.

supply coming in appears to be 2x single phase suppliers. its possible that they are also split phase 230/460v

 

[GEM Electrics]

Yes I concur with the loose connection, hence the tapering off of the discouleration, but even that would require a substantial current. It would be worthwhile looking round the night storage heaters to check their total wattage and voltage ratiing, which by law must be attached to the unit, but where and if it is still legible may be in doubt. The tails of course should be 16mm for 60A loads or 25mm for 100A loads, but if they run cold then they are probably sized correctly, but I would still replace and damaged cable.

The picture confirms the 3 phase meter, So I am worried about the possibility of 440 Volts. Be very carefull!

The RCD number you give confirms a 3 phase RCD rated at 100A (per phase) and that its operating Residual Current is 30mA. Now the problem is that the EIC advise against a single protective device that would plunge the whole house into darkness if it switched off. Added to this Ed18 of BS7671 Reg 134.2 goes even further saying that each circuit should be so protected that a fault on one one circuit shall not affect another circuit. The only way to achieve this in a TT installation is to use 30mA RCBO's on each circuit with a 100mA S type (time delayed) RCD overall. (And I know this by experiance having been taken to task by my assessor some years ago). Actually I have just looked up the 100mA Type S RCDs and the price is not that bad... depends what you call bad! they are aroudn £60 to £75.

 

[Andy™]

fitting a 100mA s type brings back the exact same problem, if that trips everything goes off. so either youve just conned the customer out of a load of money doing unnecessary work and still not fixed the issue of 1 RCD on everything, or quite simply, there is no need for the 100mA time delay since any fault would be dealt with by the RCBO's making the upfront RCD redundant...

 

[GEM Electrics]

You have missed the point entirely, re-read my post..

If the EFLI is 0.000244 amps, what will happen in the event of an earth fault?? [or more to the point not happen] Now, if a person touches an exposed conductive part in these circumstances, it does NOT mean that the current that flows through THEM will be 0.000244 Amps. It will be more like 0.25 of an amp... What will happen to them??

john..

Hi John,
Sorry if I have missed your point. I try not to do that.
What will happen at 0.25Amp? the RCD will blow within 40mS. Seems too obvious, what have I missed this time?
Geoff