Hello everyone,

I've been meaning to make a post on this for a while, but hadn't gotten around to it.  I can't wait to hear your thoughts on this one.

Anyways, a while back I noticed something peculiar with our home's electrical system.  We live in an older home (built in 1953), which still has all the original wiring, except for some circuits added in 2006.  Like any old home, the wiring isn't good by any stretch of imagination - undersized wiring (for the breaker size), no ground conductors, and over-extension of circuits is just the beginning.

However, I've never seen what I'm about to mention here.  It turns out that one of our circuit's "neutrals" functions by being grounded to a water pipe.  How do I know?  Let me explain.

I first started to suspect something was off when I noticed my Honeywell air filter (or any appliance with a motor) would suddenly quiet down (as if there were a drop in current), whenever someone would flush the master bathroom toilet.

Also, one day I was running a space heater on one outlet in my room (on the master bedroom circuit).  Under my desk, I had a wall-mounted surge protector with ground status LEDs, on an entirely separate circuit.  What puzzled me was that when the space heater (on the other, separate circuit) was on, the ground LED on the surge protector (NOT on the heater's outlet circuit) would inexplicably dim.  But, when I turned the heater off, the light went back to its normal brightness.  To test this, I even turned the heater to max, watched the LED on the surge protector, and turned the heater off one notch at a time.  The ground light flickered back to full brightness!  But, the surge protection light never flickered.

So, not only is most of our home's electrical system grounded to plumbing, but we have a neutral fed off a water pipe, too.  Luckily, the house hasn't burnt down, but to think the wiring has been that way since the house was built is unbelievable!  How could no one have ever noticed this?!

I know many people on here will say we should re-wire, but my parents detest house projects and don't see any issue given how long it's been this way. 


Hope you like!

There is important to distinguish between PE (protection grounding) vs N (intended too carry working current).
For the first is supposed to be interconnected to all the piping as well, in TT systems (PE has separate grounding from N) it could be just the piping providing this connection.
In TNC (where PE and N share common conductor, then called PEN) the PEN is grounded on multiple places along its length.

What could be your problem:
- The installation is made as TNC (was that ever code-legal in the US for homes?) and the PEN broke on some place and so the currents start to fliw through the grounding piping.
- Someone had made a mess in the wiring and "identified" and then connected whatever conductor had no voltage on it as N, so connected the lad N to a PE wire (grounded only via the piping, which was OK for the PE line; of course not for the N).

Definitely this mess is dangerous not only because of fire hazard, but because of electrical shock coming from a water faucet or any metal structure outside of your house - depends where the voltage drop is...
And it signalizes, the water piping is in questionable state too (high risk of water leak, where the voltage drop is - the passing current may have compromised the seal there).


Using ground as a working conductor was legal only for HV SWER lines (single wire earth return; a cheap way to supply very few homes over long distance - you pay just one conductor, plus because tgere is no other to short to when swaying in the wind, poles could be spaced further apart, reducing the cost further), but first this always ended on a transformer pole and the current return always used separate grounding electrode (spaced away from the LV side PE grounding) and because of the HV (10's kV) and just few kW, the grounding currents were very limited (barely an amp), so did not cause tyat much voltage drop on the grounding electrode.
But as far as I know, SWER installations were legal only for really remote houses and even for at least a decade new are not allowed anymore in the US (you need a metallic conductor for the return line, but that could be way closer to the ground, so the spacing advantage, as well as the ability to use cheaper steel is still there).

Like !

For what i know, in the US the point of separation of PEN is the busbar in the panel from which the 15/20A circuits go. (This is in contrast to European panels, where the bars are separate and the point of separation is more upstream, although it can still be a terminal in the same panel). But i guess that in a subpanel there would be already separate PE and N coming from the upstream panel

This could have lasted for years for any of a few reasons :

1. The fact that the return path is electrically disturbed by starting and stopping of the flow of water in the pipe, means that the circuit may be closed in some point through the water. This also means that parts of the plumbing are acting as electrodes (on the inside surface, facing the water). Maybe they have deteriorated over the years

2. Maybe a section of nonmetallic plumbing have been installed since. (Although i think its unlikely to be as simple as that : Any length of water column in a domestic plastic pipe have by far too high resistance to provide even remotely functioning N for higher powered appliances. The space heater would not work and lights would dim to near zero when it is switched on)

3. There have been other parallel paths for the return current, that have been broken. For example : Some metal structures or pipework inside and outside the house, installation/removal of an unintended remote PE source by e.g. cable TV)

4. Higher current draw nowadays than in the past on the same bad connection

I know many people on here will say we should re-wire, but my parents detest house projects and don't see any issue given how long it's been this way. 

Re-wirings are awful when wires are old, but i would definitively call a qualified electrician on the spot as soon as possible:
- if is true neutral conductors are connected via water pipes, in case of failure of main neutral conductor (or replace some pipes with non metallic pipes), water tap will become live; maybe lack of ground connection is saving your parents when they touch metallic appliances.
- if there are undersized conductors, they are a fire waiting to happen (due wire heating).

Like !

For what i know, in the US the point of separation of PEN is the busbar in the panel from which the 15/20A circuits go. (This is in contrast to European panels, where the bars are separate and the point of separation is more upstream, although it can still be a terminal in the same panel). But i guess that in a subpanel there would be already separate PE and N coming from the upstream panel

Well, in Europe the point of separation is dictated by the conductor cross section: TN-C is allowed only for sizes above certain limit (if I remember well, it is 16mm^2, but my memory could be wrong on this exact number; which is way above what typical homes use, so TN-C goes along the streets, but branches to homes are then TN-S).
It seems to me in the US very similar rule is in effect: The line till the panel is above the limit so may be common, the branches are below, so it is separated.
And maybe other requirement could be an assumed the balanced load cancels out the bulk (except the "last" 15..20A) of the Neutral current: If the N current does not cancel out, it can not be common with PE. I see this rule being followed in real installations, but I dont know if it is really explicitely required.

This could have lasted for years for any of a few reasons :

1. The fact that the return path is electrically disturbed by starting and stopping of the flow of water in the pipe, means that the circuit may be closed in some point through the water. This also means that parts of the plumbing are acting as electrodes (on the inside surface, facing the water). Maybe they have deteriorated over the years

I doubt the water itself could carry any significant current. See the electrolytic heater based steam generators (two electrodes, with the water fill becoming the heating element; it means it inherently can not overheat when running dry): You have to add salt for them to work at all.
The effect I would expect to be a tube joint is overheating when the water is stopped, but gets cooled down when it is running along again. These temperature changes are then what causes the changes in conductivity of the joint.


Anyway the installation state is really sketchy and needs to be fixed ASAP, it is really a disaster waiting to happen very soon...