Theory question about neutrals in circuits

[ffp]

Ok, first I want to say that I am an electrical engineer. I want to be very clear that I am trying to understand a specific situation that I came across. I am searching for an answer of why and how this situation could happen. I don't want a solution for the problem, just a theoretical, physical explanation. The other thread I posted was locked due to an apparent "safety" issue. I can't stress this enough, I am not looking for tips about what to do or how to proceed. My issue is with the theory behind this. So we might as well treat it as an hypothetical situation.

This is an example circuit of a 2 phase house (220 V phase-neutral). Phase C of the transformer goes to another house, while phase A and B are the phases of this house.
C1,C2,C3 and C4 are the 4 circuits this house has each with its respective circuit breaker. N is the neutral bar that is shared among all circuits in N1 and N2. Tomada 1, 2 and 3 are the outlets 1, 2 and 3.

Now, the neutral opened right after the neutral bar in N1, affecting C1 and the two outlets that belong to this circuit. If you put a load in outlet 1, the potential of phase A will be transferred to N1 that is connected in the outlets. However, there will be no voltage because there is no potential diference and a multimeter in outlet 2 would register 0 V.Any device pluged in outlet 2 won't work. Outlet 3 will work normally.

Now, consider this last scenario, where the opening happens before the neutral bar. Now the load in outlet 1 also transfers the potential of phase A to N1, there is no voltage in outlet 2 and there will be a higher voltage in outlet 3, since it's neutral now has the potential of phase A. Outlet 3 is now a phase-phase 380 V outlet.

Now, I would like to question:

1- If I say that the load plugged in outlet 1 in both scenarios is still working. What could possibly be allowing this to happen?

2- If there was a combination of the two scenarios, an opening before the neutral bar and an opening in both neutrals (N1 and N2), and the multimeter in outlet 2 registered 220 V. How could that be possible?

 

[Baluncore]

1- If I say that the load plugged in outlet 1 in both scenarios is still working. What could possibly be allowing this to happen?

Somewhere in the house, N1 and N2 are connected, or have been confused so outlet 1 is in series with another load, between the two phases = 380 V.

2- If there was a combination of the two scenarios, an opening before the neutral bar and an opening in both neutrals (N1 and N2), and the multimeter in outlet 2 registered 220 V. How could that be possible?

Again, there is a connection between the neutral N1 and N2, or the wrong neutral has been used for some outlet load. A high impedance voltmeter is an unreliable detector of AC voltage, unless it is in parallel with a load.

 

[DaveE]

In my experience, when you are faced with a problem that appears paradoxical, you need to go back and very carefully examine your assumptions. It is usually the case that the question you think you are solving isn't, in fact, the question you need to solve. We are often better at solving than observing.

 

[ffp]

Somewhere in the house, N1 and N2 are connected, or have been confused so outlet 1 is in series with another load, between the two phases = 380 V.Again, there is a connection between the neutral N1 and N2, or the wrong neutral has been used for some outlet load. A high impedance voltmeter is an unreliable detector of AC voltage, unless it is in parallel with a load.

That's what I thought. However, imagine now that the house has 7 different circuits (7 circuit breakers) and this situation still happens. This means all 7 neutrals are connected? I believe this is impossible, since the circuits paths are not all of them interconnected in the house, except for the neutral bar. Unless there is a connection of all of them right after the bar, where all the 7 neutrals go down the conduit.

In the case of a connection between N1 and N2 or a wrong neutral, wouldn't there be a higher voltage there? Even with no opening in the neutral?

 

[Baluncore]

In the case of a connection between N1 and N2 or a wrong neutral, wouldn't there be a higher voltage there? Even with no opening in the neutral?

No. The neutrals are all at the same potential.
You show a third pin on the outlets, but have assumed that the Earth has nothing to do with the fault. I expect one link from the ground bar to the neutral bar in the distribution box. If ground and neutral are shorted on a chain, a break in neutral will have no effect. An outlet will still operate if Earth and neutral are swapped, no matter how many breaks there are in the neutral. It is only when an RCD is installed that the EN short or EN swap would be detected.

 

[ffp]

No. The neutrals are all at the same potential.
You show a third pin on the outlets, but have assumed that the Earth has nothing to do with the fault. I expect one link from the ground bar to the neutral bar in the distribution box. If ground and neutral are shorted on a chain, a break in neutral will have no effect. An outlet will still operate if Earth and neutral are swapped, no matter how many breaks there are in the neutral. It is only when an RCD is installed that the EN short or EN swap would be detected.

I know, but now you have 2 different phases in the circuit, since neutral of phase A and phase B are connected. There can't be 220V, right?

And there is no ground in the house. The only ground there is, is located in the concessionary transformer that uses a grounded Y conection.

 

[Baluncore]

I know, but now you have 2 different phases in the circuit, since neutral of phase A and phase B are connected. There can't be 220V, right?

But that is the normal situation. All the neutrals are connected together at the neutral bar.

 

[ffp]

But that is the normal situation. All the neutrals are connected together at the neutral bar.

Oh you're rigth. But then, when they are connected i believe the grounded neutral of the transformer gives the 0 potential. Now, in the situation where there is no neutral coming to the bar, that suposedly internal connection of the neutrals N would close the circuit, but at a higher voltage, right? How could an outlet register 220 V?

 

[Baluncore]

How could an outlet register 220 V?

The series connection of two loads on two phases can generate about 220V.
It all depends on the load impedance and how you measure the voltage.

 

[ffp]

The series connection of two loads on two phases can generate about 220V.
It all depends on the load impedance and how you measure the voltage.

What if there is no load in that circuit except for maybe some light bulbs?
Also, the multimeter would have to enclose both (or more) devices to see the full voltage reduction. If it is just put on the phase and neutral terminal of an outlet of this circuit that only has loght bulbs, it would show 380V, am i right?

I don't get how a circuit with 3 outlets and a couple light bulbs would not show 380V in that case...

 

[Tom.G]

To help sort it out, try disconnecting (at the breakers?) various loads individually and in combination.

With good documentation of measured voltages and many sketches, the answer may well pop out.

 

[Baluncore]

What if there is no load in that circuit except for maybe some light bulbs?

Light bulbs are a real load.

If a wire is open circuit, you may be reading the capacitive coupling between the wires. The problem is when you measure a voltage with a modern multimeter you are drawing insufficient current to get a true voltage reading. You need to momentarily connect a load such as a filament bulb in parallel with the meter to identify how reliable the voltage reading really is.

 

[ffp]

To help sort it out, try disconnecting (at the breakers?) various loads individually and in combination.

With good documentation of measured voltages and many sketches, the answer may well pop out.

I believe that by doing that what eould happen is that the circuit of the breaker turned off will turn off and its outlets register 0 V. The others will keep registering 220 V. Don't know how this would help.

Definitely will sketch the whole circuitry later..

 

[ffp]

Light bulbs are a real load.

If a wire is open circuit, you may be reading the capacitive coupling between the wires. The problem is when you measure a voltage with a modern multimeter you are drawing insufficient current to get a true voltage reading. You need to momentarily connect a load such as a filament bulb in parallel with the meter to identify how reliable the voltage reading really is.

It is no longer opened when i insert the multimeter. And in voltmeter, the resistence is very high. I don't understand your point.

 

[Baluncore]

I don't understand your point.

When you read voltages with only a high impedance meter you can fool yourself.

Lay a wire through a building and do not connect it to anything. Measure the AC voltage on the wire relative to the Earth or neutral. It will have a voltage due to capacitive pickup from live wires in the building. Now connect a filament globe across the multimeter and the voltage reading will fall close to zero.

 

[Tom.G]

I believe...

That's why I suggested DOING it!

 

[Svein]

In my experience (coming from a country with predominantly three-phase delta power line distribution) you do not trust "neutral" to be "safe". I always treat both wires as if they were "live".

 

[ffp]

When you read voltages with only a high impedance meter you can fool yourself.

Lay a wire through a building and do not connect it to anything. Measure the AC voltage on the wire relative to the Earth or neutral. It will have a voltage due to capacitive pickup from live wires in the building. Now connect a filament globe across the multimeter and the voltage reading will fall close to zero.

I am sorry. I think I still don't understand your point. What you mean by filament globe? Are you saying that the voltmeter is not reliable? If that's what you're saying I disagree. For the levels of voltage found in ressidential electric systems it is pretty accurate. I never had trouble reading voltage this way.
If I misunderstood your point, could you elaborate?

 

[ffp]

That's why I suggested DOING it!

What kind of behaviour would you expect from what you said?

 

[berkeman]

I am sorry. I think I still don't understand your point. What you mean by filament globe? Are you saying that the voltmeter is not reliable?

I believe he means a standard incandescent light bulb, like a standard lamp. It is a relatively safe way to tell if that high voltage you are reading with your high input impedance DVM is really AC Mains or not. If it is capacitively coupled noise at that level, your meter won't be able to distinguish that. The light bulb can...

 

[Nugatory]

In my experience (coming from a country with predominantly three-phase delta power line distribution) you do not trust "neutral" to be "safe". I always treat both wires as if they were "live".

A digression in this thread, but even in North America (120/240 single phase, neutral ostensibly bonded to earth) neutral should be considered live. It’s much less likely to be dangerous, but it only takes one switched-neutral or loose-neutral experience to demonstrate the difference between “less likely to be dangerous” and “safe”.

 

[Baluncore]

Ok, first I want to say that I am an electrical engineer.

I am sorry. I think I still don't understand your point.

What make and model of voltmeters do you use ?

 

[artis]

Look @ffp normally in a three phase system if one wants to have only single phase like in an apartment one only uses one phase wire and the neutral for the whole apartment. It is never good practice (or even allowed depending on the place) to use two or three phases for different wall sockets within a single flat. Why?
Simple. Because even if everything is functioning properly with the wires instead of having a maximum voltage of 220/230 to 0 you now have 380/400 between say two wall outlets each connected to it's own phase.In your case I am not sure is that schematic representing a real situation or just a thought in your head (something tells me it's real) but anyway you are having 3 sockets of which 2 are connected to one phase while the third to another, the neutral is shared across all 3 of them but since it is disconnected from the transformer , it now becomes like a separate wire connecting the outlets in series. In your case your load (carga) in outlet one is now in series with your second voltmeter in outlet 3, since they are connected in series across two phases the 380 volts would be roughly divided in half so your second voltmeter would not show 380 but rather a lower voltage something closer to 220.
As an electrical engineer I'm sure you already know that voltmeters have high internal resistance/impedance so the total current running through such a connection would be very very small.

To answer your questions 1 and 2 the outlets could only still work if the neutral was tied to ground/neutral somewhere along the way not shown in your schematic, the other option as in your last schematic is that there is a load connected in socket 1 and 3 which then makes a series connection between phases A and B.

 

[Averagesupernova]

Look @ffp normally in a three phase system if one wants to have only single phase like in an apartment one only uses one phase wire and the neutral for the whole apartment. It is never good practice (or even allowed depending on the place) to use two or three phases for different wall sockets within a single flat.

What you describe here is common practice and at times a requirement in the USA. However, the voltages are 120 from any phase to neutral for residential. Apartments served with 3-phase likely have 208 volts between some outlets.
-
Residences (houses) will always have 240 volt single phase fed to them. The neutral is the center tap of a transformer secondary and is grounded. It's not uncommon to have two separate circuits in the same box (double gang switch for instance) that are opposite phases. Probably more properly referred to as legs in the single phase scenario.

 

[artis]

@Averagesupernova yes I agree the US has those two split phases but as you said they equal to 240v which is standard single phase voltage here in EU. I don't know how it is elsewhere but at least here it's not common practice to wire separate phases into the same apartment to different rooms/sockets because we use 230/400 volt system and someone might touch a surface that is leaking and put into one phase socket while then grabbing something from a socket with a different phase and get zapped with 400v.
Also there might be equipment malfunctions if they share some live circuit element yet are put into different phases.

If one needs three phase equipment we have a separate 3 phase socket for that.

But as I said I can't speak for the whole world I only know the places I know.

 

[anorlunda]

Any and all discussions of home wiring get confused on PF. We have members from many countries, each of whom sees things from their local reference. Local practices vary considerably, country to country.

In one old thread (sorry, I can't find it any more) someone posted a chart showing 170 (if my memory is correct) different wiring standards.

These topics would be best handled on single-country forums. PF is not that.

 

[berkeman]

Thanks @anorlunda that is a good point, and it is a good time to close this thread. Thank you everybody for trying to help the OP.

 

[berkeman]

Thanks @anorlunda that is a good point, and it is a good time to close this thread. Thank you everybody for trying to help the OP.

After an extended discussion with the Mentors, Advisors and the OP, this thread is reopened for now. The OP has been advised to hire a licensed electrician to figure this out in his location, but so far that suggestion is being resisted...

 

[ffp]

I believe he means a standard incandescent light bulb, like a standard lamp. It is a relatively safe way to tell if that high voltage you are reading with your high input impedance DVM is really AC Mains or not. If it is capacitively coupled noise at that level, your meter won't be able to distinguish that. The light bulb can...

Hm. I might be wrong, but I believe that to result in 220 V by capacitive coupling, there should be a higher voltage on the wires resulting in a higher electrical field. The wires are insulated inside the conduits. Also, to result in an exactly 220 V by capacitive noise would be a really strange coincidence.

What make and model of voltmeters do you use ?

It is a national brand from here. You probably won't recognize it. However it is a good multimeter and I have used it several times with no issue. I don't think the problem is the multimeter...

Look @ffp normally in a three phase system if one wants to have only single phase like in an apartment one only uses one phase wire and the neutral for the whole apartment.

Sure, if you just want one phase, you only bring one phase and the neutral. For a two phase system, we bring two phases and the neutral.

It is never good practice (or even allowed depending on the place) to use two or three phases for different wall sockets within a single flat. Why?

Simple. Because even if everything is functioning properly with the wires instead of having a maximum voltage of 220/230 to 0 you now have 380/400 between say two wall outlets each connected to it's own phase.

A two or three phase system will use those phases in different wall sockets. The reason for using more than one phase in a house is because the load of the place is high and having only one phase would result in a very high current in the circuits. So, we divide those loads between the teo or three phases in a way that the load is balanced.

In my OP, the system is two phased. This means that there are outlets with different phases.
Now, if you're talking about using two phases in the same outlet, well, we do that to achieve exactly a higher voltage that some devices require. In my own house I have 2 or 3 outlets with 220 V phase-phase (in my house phase-neutral is 127 V). The AC devices here work at 220 V, so we have to use a phase-phase outlet for them (220 V).

In your case I am not sure is that schematic representing a real situation or just a thought in your head (something tells me it's real) but anyway you are having 3 sockets of which 2 are connected to one phase while the third to another, the neutral is shared across all 3 of them but since it is disconnected from the transformer , it now becomes like a separate wire connecting the outlets in series. In your case your load (carga) in outlet one is now in series with your second voltmeter in outlet 3, since they are connected in series across two phases the 380 volts would be roughly divided in half so your second voltmeter would not show 380 but rather a lower voltage something closer to 220.

That is a very good observation and you're right. However, in the schematics I posted, they seem to consider the load as very very low resistance, almost a short circuit. Which is wrong in a real life scenario.
I will post a circuit that would better represent the situation so we can analyse better.

As an electrical engineer I'm sure you already know that voltmeters have high internal resistance/impedance so the total current running through such a connection would be very very small.

I do. But we are measuring voltage, not current.

To answer your questions 1 and 2 the outlets could only still work if the neutral was tied to ground/neutral somewhere along the way not shown in your schematic, the other option as in your last schematic is that there is a load connected in socket 1 and 3 which then makes a series connection between phases A and B.

Then it wouldn't be necessary to have a ground/neutral somewhere.. I mean if the circuit is closed by the loads. But without any load (suppose there aren't any device or lightbulb on) the only explanation for having a voltage in any outlet is a "ghost" neutral connecting somewhere in the circuit.
Ok, I made a more realistic version of the circuit. This is still NOT the real circuit. It's still a thought experiment, trying to understand the theory of it:

In this circuit, we have 2 phases (the AC generators) and 4 brakers. The resistances are devices or just light bulbs (please don't mind the 4.7K ohms, I forgot to change that). The open wires are the outlets of the circuits that have no device plugged in. I know they can be ignored in circuit theory, but they might be relevant here, to show exactly where along the circuit I am measuring voltage.
the big line in the right is the neutral bar where the circuits connect.

So, in this case, we have several resistnces in parallel that could be turned into a single resistence for simplicity. In this case they would be indeed resistence in series in each breaker and even for different phases. However, when I measure the voltage, I do it in an outlet, that is in parallel with just one set of resistances (each of them also in parallel) and not across both phases.

So, I might have confused myself with this, but wouldn't the voltage drop due to the resistances being in parallel not be measured if i measure the voltage in one of those "open circuit" outlets?PS: In this drawing, the company neutral is still connected. Please consider that it isn't for the questions.

 

[ffp]

After an extended discussion with the Mentors, Advisors and the OP, this thread is reopened for now. The OP has been advised to hire a licensed electrician to figure this out in his location, but so far that suggestion is being resisted...

I am grateful for you re-opening the thread as I asked. Thanks. However, as I said before, I have some points about the electrician: the house is not mine is my friend's, he did called an electrician and the guy told him to change all the wires and circuitry of the house, lastly I don't want to solve the problem, but understand it. I know no electrician that will have the amount of knowledge that people here have, specially in my country.

@anorlunda I have provided al the information about my country electrical system: voltage, transformer secondary connection (Y with grounded neutral). If you or anyone need more information about it, I can provide.
There is no need to bring any country details, if there is something that is needed to know I can provide that information so we can work to understand the theory behind the issue.

 

[berkeman]

@anorlunda I have provided al the information about my country electrical system: voltage, transformer secondary connection (Y with grounded neutral). If you or anyone need more information about it, I can provide.
There is no need to bring any country details, if there is something that is needed to know I can provide that information so we can work to understand the theory behind the issue.

In the US, the NEC (National Electrical Code) is the law. Can you please provide a link to the National Electrical Code laws in your country? Thank you.

https://en.wikipedia.org/wiki/National_Electrical_Code

 

[ffp]

In the US, the NEC (National Electrical Code) is the law. Can you please provide a link to the National Electrical Code laws in your country? Thank you.

https://en.wikipedia.org/wiki/National_Electrical_Code

Hm, I live in Brasil. We don't have a single document with all the laws and norms of electric system. We have a federal company called Aneel, that regulates almost everything regarding electric systems. If you tell me what exactly you want to know I can search and post what you need to know.

However I feel that your question lies in a worry that I might be doing something illegal why else would you want to know laws instead of characteristics of the sytem? Is that it?

 

[berkeman]

Hm, I live in Brasil. We don't have a single document with all the laws and norms of electric system. We have a federal company called Aneel, that regulates almost everything regarding electric systems. If you tell me what exactly you want to know I can search and post what you need to know.

However I feel that your question lies in a worry that I might be doing something illegal why else would you want to know laws instead of characteristics of the sytem? Is that it?

In the US, depending on the county, we need to file a permit application with the local building department when we want to make changes to the electrical system in houses or businesses. This is for a good reason -- it is to be sure that changes to the electrical distribution panels or other major upgrades are safe, and comply with NEC standards.

I once contracted with a company to install an all-house fan in my home, and they tried to attach that high-power device to an existing circuit and breaker. When I saw that, I pointed out that the NEC required that they pull a local permit and install a new breaker for that circuit, and they finally complied.

It is similar to the local US building code requirement that you pull a permit and have inspections when you modify load bearing walls. When you make additions to the electrical distribution in a home or modify load bearing walls, you may endanger other people's lives if you do it even a little wrong.

I'm not sure that trying to justify this discussion based on your claim that you are in a 3rd world country that does not require safety standards and inspections and permits is valid. Thread closed again. Lordy.