Losing neutral in the utility system

[Averagesupernova]

You mention Earth resistance. Between where and what? You have a fault someplace that is leaking current off to where it should not be? Do you not connect your neutral to ANYTHING over there besides the wire that heads to the transformer? There is no connection to municipal water system? I have had conflicting replies to that question.

 

[sophiecentaur]

Do you not connect your neutral to ANYTHING over there besides the wire that heads to the transformer?

No. Never, afaik. The supply company gives you Neutral and Live tails plus an Earth (when they have an armoured outer). Including any Earth connection in the supply downstream of the RCD breaker will cause it to blow. Aamof, connecting the neutral on a (disconnected) circuit in the house will cause the RCD breaker to cut out, which can be inconvenient when you are trying to use the house lights at the time!

 

[Averagesupernova]

No. Never, afaik. The supply company gives you Neutral and Live tails plus an Earth (when they have an armoured outer). Including any Earth connection in the supply downstream of the RCD breaker will cause it to blow. Aamof, connecting the neutral on a (disconnected) circuit in the house will cause the RCD breaker to cut out, which can be inconvenient when you are trying to use the house lights at the time!

I am still confused. Post #14 contradicts what you have said. To me it does anyway.
-

The incoming Neutral is Earthed at the sub station, at the centre point of the transformer. This then provides the E connection for everything at the house in addition to being the N conductor. An E wire (third wire) is provided to every socket. There is no actual ground connection at the house, apart from bonding of pipes and metalwork. So the difference between US and UK on the diagram that N is connected to the service panel metalwork. If L touches the metalwork of an appliance, current flows via the E wire back to the service panel where it then flows via the N (armouring) of the incoming cable. So there is no requirement for a Ground Fault Interrupter (Residual Current Device), although such devices are very commonly used for each ring main.

The armouring on the cable I am to understand is the actual neutral and not a third conductor. But that contradicts your above quote. So if fault current were to flow on this conductor it implies there is a connection to the panel chassis which in the diagram in post #11 does not show to exist.
-
Post #16:

The cable is of "coaxial" construction, with two conductors, a central L conductor and a steel armouring as the outer. The outer provides the N and E path. At the house, E and N house wiring is connected together at the entry point.

-
I just want to understand it. Not trying to say so-and-so is wrong and the other guy is right.

 

[Averagesupernova]

While the creation of this thread was to discuss wiring in the USA it has taken a turn to discussing wiring and grounding methods across the pond as well. A quick google turns up a wiki article:
-
https://en.wikipedia.org/wiki/Earthing_system
-
It seems there is no hard and fast standard. There is some new terminology to learn for those of us in the USA.

 

[sophiecentaur]

As far as I know, the Neutral and Live come directly from the substation transformer. If there is an underground connection, that will have a steel armoured sheath which is 'Earthy'. Without exception, in my experience, there is always a volt or two on the Neutral (with respect to Earth) and they are not connected together locally. This, I think is because the usual urban system in the UK supplies three phases and neutral along every road (separate armoured outer). Every third house gets one of the phases and the neutral. This arrangement allows the neutral to float around the 'middle' point of the three phases. It is, afaiaa, forbidden to connect the neutral to Earth in your house because it would interfere with this floating situation. I have no company Earth conductor and there must be a local Earth somewhere. I hope it is more than just the water system. (I can't see any Earth bond at the gas meter, which is a surprise to me. Watch this space.
The Earth is used as a safety measure and that was all there was, until recently. Nowadays we have RCDs to detect L/N current imbalance. The original 'old fashioned' Earth leakage trip was used to detect leakage to Earth if the Earth resistance was too high to allow a fuse to blow whilst an appliance slowly fried. It used to detect Earth fault current by comparing the volts at the domestic Earth (taking the fault current) with a local Earth reference taking no current. Not a biological safety feature - just a way to prevent fires etc..
UK and US are a million miles apart as far as electricity is concerned.

 

[jim hardy]

It is, afaiaa, forbidden to connect the neutral to Earth in your house because it would interfere with this floating situation. I have no company Earth conductor and there must be a local Earth somewhere. I hope it is more than just the water system.

Hmmm interesting.

In US we have to provide a low impedance all metal path for fault current back to the transformer winding from which it came. That path is to be independent of Earth ground, because Earth is likely a few ohms not the fraction of an ohm you get from an all metal path. It is also independent of Neutral, the return path for load current.* see edit below...
That path can be wire, metallic conduit, or metal raceway.
That path has recently come to be called in our electrical code the "Bonding conductor" , and it carries fault current back to its source, the service transformer,
as opposed to the "Grounding conductor" which ties the bonding conductors (and neutral conductor) to Earth ground and should carry only small amounts of current.

The revised terminology has caused much confusion in US electrical code circles.

*edit it is still common practice though, on utility side of meter, to use a single conductor to provide return path for both both Fault and neutral currents. My house has three wires coming in from pole to meter. But i must run four conductors to my new guest house, two hots a neutral and the equipment bonding conductor. Since i ran plastic conduit out to it i ran a fourth wire for bonding .

You Brits can easily get by with three wires since you have only one 'hot', i'd think. But where's your low impedance fault return path in a two wire feed ?

old jim

 

[jim hardy]

as i said there's a lot of confusion stateside.

http://www.mikeholt.com/instructor2/img/product/pdf/1292432628sample.pdf

 

[sophiecentaur]

The main need for a 'good earth' is to ensure that all touchable conductors are held at the same potential, even when there is a fault current flowing. Bonding of everything is far more important than having a good connection to an Earth back at the substation because no one can get a shock if there is no PD.
My situation (Just two overhead conductors from the Company*) is pretty unusual. Nearly everyone in the UK gets an underground feed with an armoured sheath (earthed at the sub station).

In US we have to provide a low impedance all metal path for fault current back to the transformer winding from which it came.

I realize how desirable that is but how can the consumer make that happen? They only have access to the equipment on their premises. Can you always be sure that the supplier will give you an Earth at the edge of your land?
Jim. You know about these things. Am I right about the usefulness of having a neutral that can float as the three loads vary independently on a three phase system (as in UK housing)? I think it allows the input phases to the substation transformer to be better balanced, even when the outputs are not. Is that nonsense? The requirements are quite specific that consumers must not bond Neutral and Earth together.

*My supply arrangement makes me cringe. There are poles along the road with saggy cables from each pole to the eaves of two houses per pole. The lines are low enough to touch with long handled garden tools. They are insulated of course but it just looks so dodgy! And the catenary swings alarmingly about in the wind. One day I swear it will give up and fall to the ground. tzzz. tzzzz. tzzzzzzz! But life has its compensations. No street lamps and very few cars, except for one hour am and one hour pm. Peaceful. (Just us and the rats)

 

[Averagesupernova]

Am I right about the usefulness of having a neutral that can float as the three loads vary independently on a three phase system (as in UK housing)?

Isn't the neutral earthed at the substation/transformer in the UK regardless of whether there is an actual conductor between the house and transformer dedicated to fault current? How can the neutral float?
-
About your concern of whether the power supplier gives the consumer an Earth at the property edge in the USA. The neutral is the earth. They are the same conductor. Fault current will be on the neutral. This is the whole point of this thread. The fact that this shared conductor is grounded to water systems and etc. greatly increases the chances of multiple paths for neutral current.

 

[sophiecentaur]

How can the neutral float?

The neutral volts are never zero.That's what I mean by floating.
There is finite resistance in the neutral line form the transformer - especially to the end of the line. The PD at the end will be the vector sum of the three phase currents times 'some' resistance. It's easy to measure the RMS value of the neutral volts (to Earth) and, as I've already pointed out, it can be more than a couple of volts (=+/-1%). If the loads are perfectly balanced, the neutral current would be zero and, I guess, so would the neutral volts - but it seldom is, because I always find a finite reading, in all the houses I have looked. It happens in different houses, at different times of year (loads) and with different meters, so it's real.
But there must be a reason why suppliers forbid neutral - Earth connection by the consumer. I am merely suggesting some possible reason. Could there be something to do with the actual amount (or measurement of) Energy Supplied? A 2V difference in volts would correspond to more than 2% difference in Power (V²)

The fact that this shared conductor is grounded to water systems and etc. greatly increases the chances of multiple paths for neutral current.

Not in the UK. Current through appliances does not ever flow through any 'grounded' metal structures. It all flows in the neutral - back to the transformer. I can't understand your reasoning here. If it were feasible to use the Earth conductor(s) then why use a neutral conductor at all? But I thought Earth return was not usable in most circumstances. Think of the copper we could save!

 

[Averagesupernova]

Now don't get me wrong here. I would prefer the the system the UK uses when it comes to grounding (earthing). I have never liked multiple paths when it comes to current flow. The USA uses a dedicated wire for fault current between the main panel and any sub-panel and I think it is a good thing. The only place they allow sharing the neutral and ground is between the transformer and main panel. It's not that we want multiple current paths in the USA it is just the way it is as the system has evolved. I would say eventually the neutral and ground will be split up all the way to the transformer.

 

[Windadct]

A few things I'd like to throw in there,

For US residential the residence to residence ground-(rod) neutral bond is for the local safety and yes, now there are multiple "loops" between ground and the transformers neutral, but there is generally no load being applied in this space, and in effect, the multiple houses with ground rods create a grounding grid. So the risk of a loop problem is pretty small, and in theory help the outdoor areas in the neighborhood be grounded as well.

As for the idea of loosing a neutral on the utility side that I have not seen brought up is surge protection - in a 3 phase system a Grounded Wye (Y) will maintain V L-N on all phases - even during a fault. However if the neutral is broken, during the fault the potential from the other phases and ground will shift to L-L voltage. Probably not an issue for household surge protection, but many a pole mount lightning arrestor have been destroyed since they are chosen based on the L-N for Wye. I worked on one project where a chemical plant added a delta feed, but they used the same spec on parts as the original Wye feed - when they had a fault the Arrestors that were inside the switchgear did a lot of the damage - and the line was on a recloser - so the gear saw 3 total faults.( it was a long week)

Sophie - for a neutral - gnd fault they have a signal injector systems (http://www.greenlee.com/products/PULSE%2540dGROUND-FAULT-LOCATOR.html?product_id=18006) - but the principal is pretty creative. You can do it with a clamp on CT - injecting a pulsing current signal that can be detected. You clamp on the neutral at the bonding point, and the signal, being current will stay on the neutral until the location of the fault. You can go through the system and test each point for the signal. This can be a dc pulse - but then you need to clamp on the neutral at each location, or an RF / audible current with a detector. What is good is the use of CTs and or a non-contact detector - the process is pretty safe. I have not been able to find ones like I used to use though, at least not as cheap as I expect considering how basic the tool is..

 

[Averagesupernova]

Here is some more interesting reading:
-
http://forums.mikeholt.com/showthread.php?t=56987
-
http://www.garagejournal.com/forum/archive/index.php/t-82570.html
-
There are more, but I want to more carefully review them.

 

[sophiecentaur]

I found out something interesting today about the earthing of my property. The Company supply has only two wires, L and N (which I knew) and the measured Earth Resistance was 28Ω, according to the man with the meter. He was telling me that, until the company supply me with an Earth, an Earth resistance of up to 200Ω is within spec. Naturally, I have a 30mA RCD for protection. He also told me that 28Ω was actually "quite good" for houses in our area with overhead supplies so there would be little point in my trying to improve on it by adding another spike or net. Amazing.

 

[sophiecentaur]

Here is some more interesting reading:
-
http://forums.mikeholt.com/showthread.php?t=56987
-
http://www.garagejournal.com/forum/archive/index.php/t-82570.html
-
There are more, but I want to more carefully review them.

Those scenarios, described in the links, could not occur with an RCD in place, unless there is a short between N and E, upstream of the breaker. One guy was suggesting that a water heater could be to blame - well, not in the UK, at least if the installation has been looked at within the last few decades.

 

[Averagesupernova]

The case of the guy running his microwave oven causing a rise in current in the water and gas pipe is quite simply an indication of a poor neutral. The suggestion of the water heater didn't come very close in my opinion since the water heater would have little to do with microwave.
-
I realize that an RCD senses imbalance of current between the wires. As far as I know, in the USA the equivalent is the GFCI outlets, and sometimes breakers. No residence has the entire panel protected by a single GFCI breaker. The way we share the neutral and ground between the transformer and service panel and bonding this wire to the water pipe in the house would cause nuisance trips of such a device installed to protect the entire panel. There would be no way to guarantee all of the normal current stays 100% in the neutral conductor.

 

[sophiecentaur]

The case of the guy running his microwave oven causing a rise in current in the water and gas pipe is quite simply an indication of a poor neutral. The suggestion of the water heater didn't come very close in my opinion since the water heater would have little to do with microwave.
-
I realize that an RCD senses imbalance of current between the wires. As far as I know, in the USA the equivalent is the GFCI outlets, and sometimes breakers. No residence has the entire panel protected by a single GFCI breaker. The way we share the neutral and ground between the transformer and service panel and bonding this wire to the water pipe in the house would cause nuisance trips of such a device installed to protect the entire panel. There would be no way to guarantee all of the normal current stays 100% in the neutral conductor.

The terms of this thread are actually a bit too vague. There is a significant difference between the situation within the house (your business) and the supply (the company's business). Comments have been made which seem to confuse the two.
A "poor neutral" could be really bad news. What could make it "poor"? Possibly a single joint in the N circuit. Not a problem, in itself but L and N circuits would / should be constructed in the same way. If there were a similar bad joint in the live conductor, there would be no alternative path so you have the great possibility of a fire on the supply line. Would the Neutral circuit not be tested as rigorously as the live circuit? If there is Earth Current from a microwave oven (only), then the device is faulty, by any standards. Whatever the spec for the domestic wiring, the device would show continuity E-N. However bad the N resistance is, if the Earth conductor is properly separate, all you should expect is low volts on the appliance and a hot wire or smoke from within it.
I still don't get the point of having two alternative 'return' paths. Any appliance that takes the return current to Earth is potentially lethal. That Earth connection, to its exposed metal is there to eliminate the possibility of any danger of shock. By accepting that the return path could be allowed through its Earth conductor, you build in risk. Why is the Earth circuit not just used for safety? Does it save money? (the only possible justification for that sort of thing)
Speaking from experience, a single RCD on the consumer panel inlet has not ever caused a "nuisance trip" - except when I have been taking liberties when actually tinkering with the system. Single pole isolators can allow you to touch N and E together by mistake on a safe circuit and the RCD does trip. But I can't remember a single instance of a genuine trip. Nonetheless, there is always a lot to be said for individual circuits to be given their own breakers. That is actually the latest requirement in UK. It has been a longstanding requirement for electric showers to have individual protection.
There is another advantage with individual RCDs per circuit and that is the fact that RCDs give double pole isolation - unlike MCBs and Fuses.
I always find these UK / US electrics conversations throw up things that I had just never thought about. Ferrrry Eeenteresting (Rowan and Martin?)

 

[Svein]

I found out something interesting today about the earthing of my property. The Company supply has only two wires, L and N (which I knew) and the measured Earth Resistance was 28Ω, according to the man with the meter. He was telling me that, until the company supply me with an Earth, an Earth resistance of up to 200Ω is within spec. Naturally, I have a 30mA RCD for protection. He also told me that 28Ω was actually "quite good" for houses in our area with overhead supplies so there would be little point in my trying to improve on it by adding another spike or net. Amazing.

So, why should the "Earth" be a conductor at all? If you live in the mountains, how would you get any sort of connection to "Earth"? 28Ω resistance between a local "Earth" connection and a remote one seems almost too good to be true.

Here in Norway a connection to "Earth" is tenuous at best. Way back when the water pipes were made of copper, they would represent the best connection to "Earth" you could get, but since the advent of plastic water (and soil) pipes that connection is gone. Anyhow - since the power distribution network is three-phase delta, there is no "Neutral" anywhere and the "Earth" connection is purely for safety reasons. If you have a three-phase installation, the "Earth" is probably way off center and may even be outside the delta (I once worked at a research lab, where we did that measurement and arrived at that conclusion).

 

[sophiecentaur]

So, why should the "Earth" be a conductor at all?

That's a fair enough question. My answer is that you want to be sure that anyone even remotely associated the property should be sure that where they're standing or whatever they're holding is at the same (call it zero) potential. Also, you can never be 100% confident of an RCD working so it's nice to know that any 'short' to the building / pipes etc. will pass enough current to blow a fuse. A good Earth is the simplest, fail safe form of protection against shock, particularly the sort of shock that's just waiting to happen when someone unwittingly disconnects a conductor that is 'earthed' but passing lethal current but not enough to blow a fuse. Once the body becomes the path to Earth for that current, you have a problem.
The company requirement is for only around one Ohm of Earth resistance (when they supply an Earth connection). I feel that must have come about from experience and a lot of thought or they would surely not bother with that more expensive option.
There are more than one issue here and the thing I have the biggest argument with is the apparent preference that some people seem to have for sharing Earth and neutral as return paths.

the "Earth" is probably way off center and may even be outside the delta

Wow - creepy! You could be getting some free energy from that situation!.

 

[jim hardy]

the apparent preference that some people seem to have for sharing Earth and neutral as return paths.

just to clatrify, though i'll probably only muddy it further...

Here in US that's only permitted between the main premises entrance and the utility's transformer.
Inside the premises, with our present day code, they are separate paths joined together at a single point, where power enters the premises, as in that photo in post # 24

Because this photo is a sub-panel the green bonding screw is not engaged . That keeps neutral separate from ground at this location. They're joined only where power enters the premises and that's called "Bonding".
White striped neutral wire is the return path for load current,
Fault current returns through the metal conduit and metal of the electrical system to the main panel, where neutral IS bonded to the metal.

If no metal conduit then there's a return wire in each branch circuit for fault current that's either green or bare. My panels have another terminal strip for them that is spot-welded to the metal box. I ran a fault current return wire (four conductors total) to my guest house panel to join(bond) the metal parts together because i'd buried plastic conduit to it not metal.

Were this the service entrance adjacent meter box, the white striped wire would go out to the transformer neutral and that green screw would be engaged so as to give fault current an all metal path back to transformer neutral via that utility side neutral conductor. That all metal path parallels Earth assuring a hard fault will pass enough current to trip a breaker, and that a mild fault won't raise Earth potential out there at the fault by very much.

That's the only part of the path where load and fault current share a conductor, incoming(utility) side of main panel..
I won't be surprised if US code calls for a separate wire for fault current return on utility side of the main panel in the near future.

IEEE standard 142, the "Green Book" is a great text for introducing the concepts. IMHO all EE curricula should include a one hour course on it because the basics apply equally well to electronic circuits .

old jim

 

[Averagesupernova]

The terms of this thread are actually a bit too vague.

This thread has evolved but the original point was a weak or broken neutral between the customers panel and the transformer just as the title states.

There is a significant difference between the situation within the house (your business) and the supply (the company's business). Comments have been made which seem to confuse the two.

While there is significant difference between where you live and where I live, yes there can arise confusion. However, the companies business and my business are so intertwined that it all needs to be considered.

If there is Earth Current from a microwave oven (only), then the device is faulty, by any standards. Whatever the spec for the domestic wiring, the device would show continuity E-N. However bad the N resistance is, if the Earth conductor is properly separate, all you should expect is low volts on the appliance and a hot wire or smoke from within it.

No. This is the whole point of the thread. We are in agreement that there is a parallel path(s) for neutral current in the USA. The neutral conductor as well as the conductor that goes from the main panel in the residence to the water pipe which is connected to a water main. The neighbor who is hooked to the same transformer connects to this water main in the same way. In the USA, fault current goes back to the transformer on the neutral conductor the same way any other current does.
-
Here is a typical example of a service panel in the USA. There are 3 large conductors going into the top of the panel. Two hot wires, one each on the outside and the neutral in the middle. 240 volts between the hots and 120 volts between the neutral and either hot. Notice the smaller copper wire that connects to the same place the neutral connects. There are other heavy copper wires on the neutral bus as well. The neutral bus is on each side with many white wires and some bare copper wires. One of these heavy bare wires is the one that goes to the water pipe. If you have an all metal plumbing system to the street and your neighbor does too, you can cut the large neutral wire in the center and now all of your neutral current goes through the smaller copper wire out through the plumbing system. Normal current, fault current, doesn't matter. It's all going the same place.

 

[Svein]

Wow - creepy! You could be getting some free energy from that situation!.

Yes - but the "Earth" resistance being on the order of 200Ω (at least), you do not get very much.

Anecdote: When my new house was finished (early 2011), a guy from the power distribution company came to check the wiring. He took one reading in the fuse cabinet and exclaimed: "There is a serious ground fault here! But it is outside the house!" He reported it back and they did something about it - 11 months later. So a ground fault - almost a short circuit to "Earth" - turned out to be a minor irritation.

As to the service panel above: No way such a panel would be approved in Norway. Here is a picture of an old-fashioned service panel that will not be approved anymore.

Here is a picture of a modern service panel:

 

[Windadct]

The Electrical Industry and the codes in the USA, In My Opinion, are very heavily influenced by the electrical WORKERS - so systems that reduce installation labor are hard to get accepted. For example - the local code inspector is very often a former ( or current) electrician, and very often union, if he sees somthig that is unfamiliar - he can choose to not accept it, until the install is proven to be OK.. They are protecting the brotherhood.

Agreed, the typical US installation is decidedly Low tech - but there are other factors at play, for example 99.9% of US installs have the Revenue metering outside! If it was inside, the boogieman government would have to make a law saying the utility is guaranteed access to your household, that just will not fly here.
Theft is rampant and so when this occurs, the utility can just come, pull the meter and the house is disconnected, it is more than technical issues at play, there is a huge cultural influence.

 

[Averagesupernova]

Electrical workers often are the means that the NEC gets updated. Feedback from the field on what isn't working. This starts at the most local level. The whole thing is much more flexible than one would think. States can adopt as little or as much of the NEC as they care to.
-
I cannot agree that the typical USA installation is low tech.

 

[Averagesupernova]

Another interesting read: http://forums.mikeholt.com/showthread.php?t=180591

 

[sophiecentaur]

While there is significant difference between where you live and where I live, yes there can arise confusion. However, the companies business and my business are so intertwined that it all needs to be considered.

In the UK, my and their business have a very definite demarkation line. They deliver a 100A pair of tails (LN) from their meter to my consumer box. They may also provide an Earth. I have no influence on anything upstream and they only require some sort of sign-off certificate by the installer of my stuff. If I get no Neutral Connection the I get no power - simple as that. Earth really has nothing to do with it, as far as I am concerned.
In general, the connection between E and N is back at the supply transformer.
I get the impression that the US system is much more DIY based with much more flexibility about the installation details. It wasn't so long ago that the whole UK system was National and the Generating and Supply Boards were full of government paid 'Inspectors'.They would not be happy with the way things are done 'abroad'. But that was true with transport, food and water, too. Once Brexit has taken hold, we may be back where we started.

 

[Averagesupernova]

It varies from on company to the next here in the USA who owns what. Some places the PoCo owns everything up to and including the meter box. In some places the customer owns everything after the transformer excluding the meter itself. However, just because I may be required to own the meter box this does not imply that I have a right to open it up and tinker around with it. That part is the same as the UK in that there is a line drawn between the consumer and the PoCo. If my PoCo requires I own the hardware it means I pay for it upon installation and I am responsible to pay for replacement at the time it is deemed necessary. It sounds like the UK is more of a star ground (earth as you call it) system than here in the USA.

 

[jim hardy]

I get the impression that the US system is much more DIY based with much more flexibility about the installation details.

I think it's evolving. If i read new code correctly they want two ground rods now at building's service entrance.

The neutral needs to be earthed somewhere, and the reason is not to carry return current but to give lightning a path to earth.

My education on that was in 1979, at my brand new house i was using a drill outside when one of those sudden summer thunderstorms came up. I was downright smug, the house was wired with good grounds and i was only twenty feet from the ground rod at my service entrance. The drill was a metal housed Craftsman with three prong cord, one of Sears' best I thought i was really safe.

A lightning bolt hit the pole with a tremendous crash. At that instant the drill became a handful of fire and my feet too. I was paralyzed for about a quarter second, long enough to holler "Gaahhh", and when it let go i threw the drill. Instant education - my metal drill was connected by metal wire all the way to bottom of a lightning bolt.
That lightning bolt elevated the pole, the Earth at bottom of pole, the Earth at my service entrance ground rod, and my drill to heaven knows what potential. I experienced the ground voltage drop between my service entrance twenty feet away and my feet. Without that service entrance ground rod i'd have experienced the ground voltage drop all the way out to the pole, doubtless a whole lot more.

So Sophie, I'm sorry but i LIKE the concept of an Earth connection at building service entrance . When lightning strikes, it holds down how much the building potential elevates relative to adjacent Earth , and that's what will kill a fellow.

I think mine was a close call.
But it was sure a fast lesson. As an educator you'll appreciate that !old jim

 

[Averagesupernova]

I think it's evolving. If i read new code correctly they want two ground rods now at building's service entrance.

The neutral needs to be earthed somewhere, and the reason is not to carry return current but to give lightning a path to earth.

A single concrete encased electrode is often all that is required in place of ground rods. This involves a minimum length of rod buried in the concrete foundation of the building. In the event there was no concrete encased electrode installed, 2 ground rods are often used. I believe one rod is accepted if soil conductivity is adequate. This translates to 99% of electricians installing 2 ground rods where there is no concrete encased electrode.

My education on that was in 1979, at my brand new house i was using a drill outside when one of those sudden summer thunderstorms came up. I was downright smug, the house was wired with good grounds and i was only twenty feet from the ground rod at my service entrance. The drill was a metal housed Craftsman with three prong cord, one of Sears' best I thought i was really safe.

A lightning bolt hit the pole with a tremendous crash. At that instant the drill became a handful of fire and my feet too. I was paralyzed for about a quarter second, long enough to holler "Gaahhh", and when it let go i threw the drill. Instant education - my metal drill was connected by metal wire all the way to bottom of a lightning bolt.
That lightning bolt elevated the pole, the Earth at bottom of pole, the Earth at my service entrance ground rod, and my drill to heaven knows what potential. I experienced the ground voltage drop between my service entrance twenty feet away and my feet. Without that service entrance ground rod i'd have experienced the ground voltage drop all the way out to the pole, doubtless a whole lot more.

So Sophie, I'm sorry but i LIKE the concept of an Earth connection at building service entrance . When lightning strikes, it holds down how much the building potential elevates relative to adjacent Earth , and that's what will kill a fellow.

I think mine was a close call.
But it was sure a fast lesson. As an educator you'll appreciate that !old jim

You are a lucky man.

 

[sophiecentaur]

Eight lives to go then!
Yes. A good Earth connection between every piece of conducting metal in the vicinity is a great idea. Whether or not this Earthed network is connected to the Neutral, locally has very little relevance to the effect of a lightning strike, I reckon. The Live and Neutral have equal status when you are considering Megavolts and the live conductor is just as likely to get the sharp end of a strike.
I am not too happy about the protection I have but I can't think of any way to avoid your problem - except to give some local protection with a lightning rod in a strategic place, to smooth out the potentials around your house and reduce the likelihood of a strike in the first place.