Problem with earth indication or lack thereof.

[Artisi]

Something for the sparkies to contemplate.

Here is what seems a very strange problem / transient / conundrum whatever you want to call it either in my house electrics or influenced from an external source.

Background:

New house, 3 phase incoming, split over 5 RCBO covering all power outlets (LNE) plus 12 CB’s for other circuits including the ac units and water heaters (8 units) they are also LNE.

Problem / concern.

On one power outlet, I have a power board for the TV, DVD, satellite system etc. The power board has two indicating lamps incorporated. 1. Power-on, 2. Indicates earth disconnected if both on.

During the day only the power indicator is on, and the earth indication light is off, early evening and night the earth disconnect indicator comes on and goes off early morning, it appears that the earth disconnected light comes on at the same time as the housing estate street lights come on and goes off in the morning when the street lights go off.

Further, I have just installed a new on demand water heater in the bathroom, this has 3 lights, power, ELCB and earth, during the day the earth light is on but off during the evening / night, this also seems to coincide with the street lights coming on /off - yet to be finally confirmed but certainly appears that way. However if I power up the heater when the earth light is off, it will, but not always light up, if the light is off and I touch the water tap it will normally light.

To date there has not been any apparent problems with electrics, no tingles or bites from computers, washing machine etc. The only exception to this is, in the 2 years we have been in the house, the RCBO on the water pump has tripped twice during the night, now it’s possible that the pump wasn’t even called to come on line during the time of going to bed and getting up in the morning. It’s very unlikely to be a moisture problem with the pump / cables etc as the pump is well protected from the elements in a dry area. The pump has never trip the RCBO during normal operation.

My feeling is it is a problem coming from the street lights and the way they are connected into the power supply in the street.

Any thoughts welcomed.

[Crossy]

I suspect your installation is TT (no N-E link).

A TT system will develop a voltage between N and E according to the applied load (bigger load = bigger N - E voltage), once this voltage gets big enough it will confuse the indicator lights. These simple indicators are just that, simple and indicators, they can and do get confused under certain circumstances.

If you have a multimeter measure, record and report here: L-N, L-E, N-E voltages both during the day and in the evening when the street lighting comes on.

Can you see your ground rod, is the wire well connected to it?

Also measure L, N and E to a big screwdriver stuck in the garden (as a known good earth) at both times.

Also try connecting the big screwdriver to the earth bar in the distribution board and see if anything changes (I'm worried by the behaviour of the water heater lights).

Make the voltage measurements and let's see if you have an underlying fault or a potentially hazardous situation.

[bankruatsteve]

If your pump is plugged into an outside receptacle, take off the plate on that and clean if needed. Every time I have problem with outside receptacle, it's full of gecko shit and eggs.

[Crossy]

If your pump is plugged into an outside receptacle, take off the plate on that and clean if needed. Every time I have problem with outside receptacle, it's full of gecko shit and eggs.

It's those microscopic electricity-loving ants that get us.

Water pump, gate controller and some light switches have all died because of the little b'stards. Downstairs TV died when a gecko entered chasing after the ants.

Ant killer now in all the places they seem to like.

[Artisi]

Crossy.

Thanks for the quick reply. will take some L-N L-E N-E readings in the next day or 2 and let you know.

Can't check the earth rod connection- now buried under concrete which I am not that happy about, have been thinking about circumventing the original one and installing another earth rod with above ground connection - will put it on top of the to do list .

[Artisi]

Hi Crossy.

First lot of measurements using analogue Multimeter

Day time L-N 220V, L-E 220V, N-E maybe 1V

Night time L-N 220V, L-E 200 / 210V, N-E 27V. (street lights on)

Can't get to the earth rod.

What do you think about the 27V in the N-E ?

Thanks

Edited January 29, 2014 by Artisi

[Crossy]

OK, that explains the behaviour of the indicator lamps.

27V N-E is excessive, but shouldn't be hazardous to you or your equipment. It's normally a sign of an overloaded or badly out of balance system, little you can do about it.

As a final test to ensure no hazard, please do the screwdriver-in-the-garden (S) test, measuring S-L, S-N and S-E at both times. S-E should be low at all times, if it goes above a volt or two then you need to investigate your house earth.

I'm still concerned at the behaviour of the water heater lights when touching the tap, potentially a poor ground there, please check both ends of the heater supply for proper connection.

[Artisi]

Crossy, Thanks.

1 point though, the problem is evident without any load whatsoever, - evening / overnight the only thing/s running is the refrigerator and 2 /3 low wattage lamps.

Screwdriver test coming up, need to find cable etc to reach outside. I have very limited resources / access to necessary items here in Thailand.

Will also recheck water heater earth etc., however haven't been able to replicate light coming on when touching the tap.

One thing that maybe of interest, when the heater is on full power the earth lamp is on, when I lower the "temp." to maybe 50% of the scale the earth lamp goes off but comes on instantly when going back to full power.

[Crossy]

There's evidently some volt-drop going on, I don't think it's in your installation, more likely the village supply is on the line and not being helped by the street lighting load.

I just want to be certain that we're not looking at something we can fix, I would not forgive myself it I didn't suggest a simple test and someone got hurt as a result.

[Artisi]

Crossy, Thanks.

1 point though, the problem is evident without any load whatsoever, - evening / overnight the only thing/s running is the refrigerator and 2 /3 low wattage lamps.

Screwdriver test coming up, need to find cable etc to reach outside. I have very limited resources / access to necessary items here in Thailand.

Will also recheck water heater earth etc., however haven't been able to replicate light coming on when touching the tap.

One thing that maybe of interest, when the heater is on full power the earth lamp is on, when I lower the "temp." to maybe 50% of the scale the earth lamp goes off but comes on instantly when going back to full power.

Today second stage screwdriver test.

Daytime - L-S 210V, N-S 210V, E-S 0 V

Evening - L-S floating 200V, N-S 195 - 205V, E-S 0 V

Rechecked evening readings at 2 different GPO's, much the same as previous L-N 210V, L-E 210V, N-E 27V.

found out today we have a power authority boss a couple of doors up (boss of a different area) but will talk with him see what can be done.

[Artisi]

disregard to numbers in earlier post for screwdriver test - looks like I screwed up somewhere. will recheck tomorrow.

[Artisi]

disregard to numbers in earlier post for screwdriver test - looks like I screwed up somewhere. will recheck tomorrow.

Rechecked

The N -S reported incorrectly - read as follows.

Daytime N-S 0V.

Evening N-S 25 - 32 V

Thanks

[Crossy]

Now we have a problem, you need to check your rod is actually connected to the ground bar in your consumer unit.

Do you have a N-E (MEN) link anywhere?

You could try connecting the screwdriver to the ground bar in the CU (you were going to add a rod anyway IIRC).

[Artisi]

I'm becoming a little confused now,

1. there is no N-E link in the CU, think all earth wire/s have / are connected in the roof space and lead off to the earth rod and not connected in the CU..

2. during the day N-E is seems to be below1V, whereas in the evening it is approx. 27V. -- what should I be reading voltage wise N-E and S-E during the day / evening?

[Crossy]

N-E doesn't really matter, 27V is high but not disastrous.

S-E should be low all the time on a TT system, the fact that you are seeing a voltage between the screwdriver and your local earth suggests an issue.

Task A, since you were going to do it anyway, is to get a new rod in and link to your ground common point wherever that happens to be.

[Artisi]

nothing between S - E (measured at GPO). day or night.

Still intend talking with the power authority guy up the street, currently training Mrs. A in terminology and concern - luckily her English is excellent so can translate very easily once concept is understood..

[bankruatsteve]

N-E doesn't really matter, 27V is high but not disastrous.

S-E should be low all the time on a TT system, the fact that you are seeing a voltage between the screwdriver and your local earth suggests an issue.

Task A, since you were going to do it anyway, is to get a new rod in and link to your ground common point wherever that happens to be.

I didn't see where he got a potential between "E-S"? If there is no earth bar in the CU, then N-E, etc. testing must be at an outlet? It would be good to know where the E is actually connected to ground. I wonder if some of his lighting is connecting N-E and maybe switching N?

[Artisi]

N-E doesn't really matter, 27V is high but not disastrous.

S-E should be low all the time on a TT system, the fact that you are seeing a voltage between the screwdriver and your local earth suggests an issue.

Task A, since you were going to do it anyway, is to get a new rod in and link to your ground common point wherever that happens to be.

I didn't see where he got a potential between "E-S"? If there is no earth bar in the CU, then N-E, etc. testing must be at an outlet? It would be good to know where the E is actually connected to ground. I wonder if some of his lighting is connecting N-E and maybe switching N?

Tests have been run from a 3 pin LNE outlet.

Problem in only evident when street lights are on, irrespective of what is on not on in the house.

[Crossy]

Nah , I misread his post, combined with crappy internet today, sorry for the confusion.

Looks good with nothing between S-E at any time.

I wouldn't worry about the N-E voltage, as you're not MEN it won't do any harm, it's likely caused by things beyond your control anyway.

Since you have 3-phase it may be worth measuring each phase to neutral during the day and when the fault is evident, the overloaded phase will read low.

[Artisi]

Nah , I misread his post, combined with crappy internet today, sorry for the confusion.

Looks good with nothing between S-E at any time.

I wouldn't worry about the N-E voltage, as you're not MEN it won't do any harm, it's likely caused by things beyond your control anyway.

Since you have 3-phase it may be worth measuring each phase to neutral during the day and when the fault is evident, the overloaded phase will read low.

Thanks for the input and it does appear the cause is beyond our control. Will still take it up with the power authority for their comment, if any.

As stated initially, haven't seen / had any problem with equipment / shocks / tingles etc, except 2 "probably false" trips of the RCBO on the water pump circuit. That we can live with - if it was on the kitchen circuit with fridge etc it could be considered a problem.