Phase/neutral inversion and electronics?

[RJVB]

This should be a bit of the voodoo side of things for AC applications, but since I need some distraction from emotions:

Yesterday I discovered that the mains wall outlet in my study that powers all of my computer and audio equipment there has been wired "wrong", with the phase and neutral exchanged. (It's also the only outlet that has been grounded, and that was done properly ).

I can imagine it might make a difference with hardware on/off switches that cut only what's supposed to be the "hot" wire, but almost nothing uses those anymore. Is there any reason for me to trip the breaker and exchange the wires inside the outlet?

I've got an UPS hooked up to that outlet, for instance.

[va3ux]

Other than Code compliance plus the example you gave about switching the Hot rather than the Neutral with a single pole switch, I don't think there is any other real need to change the wiring.

[pauln]

Don't take my word for it, especially concerning Continental electric service, but I recall a convention relating to the appliance's main power fuse placement...

The AC appliances use DC for all their functions, derived from AC in the transformer, rectified and smoothed. I think the power fuse convention was that the fuse is the first thing encountered by the hot entering the chassis. I think this was to minimize damage to power supply from catastrophic failure.

[va3ux]

Don't take my word for it, especially concerning Continental electric service, but I recall a convention relating to the appliance's main power fuse placement...

The AC appliances use DC for all their functions, derived from AC in the transformer, rectified and smoothed. I think the power fuse convention was that the fuse is the first thing encountered by the hot entering the chassis. I think this was to minimize damage to power supply from catastrophic failure.

That is correct. The fuse is always in the "hot" conductor.

[thelostboss]

That is correct. The fuse is always in the "hot" conductor.

Way back in the distant past of the nineteen sixties I had to set up a couple of amps on different sides of the stage. Instruments - guitar in this case - would drive the first amplifier and an output of that amp would drive the other one. All to be done with standard instrument cables. Both amps worked flawlessly on their own but what I did not know - although I quickly found out - was that the power lead to one of the amps had active and neutral transposed.

I walked across the stage with the interconnect cable and started to plug it in, there was an enormous flash and bang, everything went dark and quiet, and I ended up about twelve feet away from the amp. The connecting cable was destroyed but the amp only suffered arc damage around the socket/jack.

The moral there is that in three wire systems everything can seem okay with active - neutral "inversion"; but that the end results may turn out to be either startling or catastrophic.

ALWAYS confirm that things are wired correctly!

After this I always carried and used a phase tester to check every wall outlet and extension lead that we were going to be using just to be safe. Never assume that commercially wired cabling is correct, we have had numerous recalls of extension leads and piggy-back plugs that introduced phase inversions.

[bluenote61]

In France you have this little pin in the socket which prevents the cable to be plugged in reverse, i.e. phase and neutral flipped.

Here in Germany we have the same socket type, but without that orientation pin, i.e. we can flip phase and neutral line each time we plug in.
And we are all still alive...

Hence, as long as the grounding of the socket and the connected device is proper made according to the technical rules, it doesn't matter on which side phase and neutral line are connected in the socket. It's all AC.

[RJVB]

started to plug it in, there was an enormous flash and bang, everything went dark and quiet, and I ended up about twelve feet away from the amp. The connecting cable was destroyed but the amp only suffered arc damage around the socket/jack.

The only time I witnessed something like that was when a colleague plugged in an old CRT terminal into some kind of robot arm controller unit, probably without unplugging both or at least turning both devices off (RS232 isn't hot-plug-safe). Actually caused the image on my screen (not 2m away) to dance as if something very heavy had fallen. He didn't get displaced though and no mains fuses were blown but just about every IC inside that controller had gone popcorn and the printed circuitry was traced on the bottom plate.

No inverted wires here - probably. Though in the Netherlands we do have the same wall outlets that don't enforce a polarity so who knows.

So there's no reason to suspect that e.g. the rectifying circuit (hasher?) in switching AC/DC PSUs will work better (and chirp less) when the wires aren't crossed?

[Ukena]

In France you have this little pin in the socket which prevents the cable to be plugged in reverse, i.e. phase and neutral flipped.

Here in Germany we have the same socket type, but without that orientation pin, i.e. we can flip phase and neutral line each time we plug in.
And we are all still alive...

Hence, as long as the grounding of the socket and the connected device is proper made according to the technical rules, it doesn't matter on which side phase and neutral line are connected in the socket. It's all AC.

There are plenty of plugs in France that don't have the pinhole – they bypass the pin altogether. Sort of like ungrounded plugs in the US, except that nowadays almost all US 2-lead plugs have a larger and smaller connector, so they can only be plugged in one way. The French 2-connector plugs are completely reversible.

Having grown up in the US and having done all my electrical learning on 110v, I was pretty used to getting those 110v shocks. I knew an electrician whose callouses were so thick he would lick his fingers to see if the outlet was live – an old school wise guy.

Now spending part of the year in a Paris apartment, I have learned about the power of continental voltage. I cut a wire that I didn't realize was plugged in – loud bang, tripped the breaker for the whole apartment, and now have a pretty big hole in that cable cutter. Very glad the handle was insulated.

I try to be a hell of a lot more careful now.

[nevershouldhavesoldit]

Reversed polarity wiring to outlets is dangerous. Devices with conductive (ie metal) parts that are connected to the “neutral” wire but are not grounded will be live with reverse polarity wiring (even when switched off, unless the power switch interrupts both hot and neutral wires. This includes the chassis on some electronics. If you touch such a part while also touching any ground (or another part at a different electrical potential relative to ground on a different device), you will get a shock.

We had this in the kitchen of our house, which we built. The builder and his subs were less than completely diligent. If I touched the metal refrigerator handle while touching the frame of the oven door next to it, I got quite a tingle. Fortunately, it was not the full 110V potential, but it got my attention and was promptly fixed. Fixing everything he screwed up took about 20 more years.

I would check every outlet in the residence to be sure it’s only the den wiring. It could be reversed at the load station rather than the outlets (also easily fixed). Whatever is wrong should be corrected, in my opinion.

[va3ux]

Reversed polarity wiring to outlets is dangerous. Devices with conductive (ie metal) parts that are connected to the “neutral” wire but are not grounded will be live with reverse polarity wiring (even when switched off, unless the power switch interrupts both hot and neutral wires. This includes the chassis on some electronics. If you touch such a part while also touching any ground (or another part at a different electrical potential relative to ground on a different device), you will get a shock.

Nothing but the ground wire should be connected to the conductive metal parts or enclosure of any appliance or device. Even though neutral is a grounded conductor, it is still a current carrying conductor and it should not be connected to any part of any enclosure or chassis.

[KirkP]

An incorrectly wired outlet is a shock hazard and should be fixed. One could debate the degree of risk, but it definitely increases risk, depending on what it plugged into it.

You should also check any other work the incompetent electrician did, since they might have made even more dangerous mistakes.

[RJVB]

Nothing but the ground wire should be connected to the conductive metal parts or enclosure of any appliance or device.

Exact, and there are strict rules about which devices should have a grounded connection and which can do without. Interestingly (for someone used to using a traditional central heating radiator as a quick "ground source"), our big electric heaters are not required to have a ground connection.

Even though neutral is a grounded conductor, it is still a current carrying conductor and it should not be connected to any part of any enclosure or chassis.

I don't know how residential 110V is produced in the US but here in Europe the (traditional) 220V (now rather 230-240V) comes from what we call tri-phasé, where local power stations would divide a single 380V line into 3 220V lines. Exactly how the current alternates between the phase and neutral poles in one of those lines would depend on the power draw in the other 2 lines, meaning that "neutral" doesn't always mean 0V.

It's actually quite likely that I replaced/installed the actual plug myself; we had to keep the electrician bill as low as possible.

[KirkP]

Nothing but the ground wire should be connected to the conductive metal parts or enclosure of any appliance or device. Even though neutral is a grounded conductor, it is still a current carrying conductor and it should not be connected to any part of any enclosure or chassis.

There are other factors that make reversing hot and neutral unsafe.

Lightbulb sockets are hazardous if the threaded part of the socket is on the hot side of the circuit. The polarized plug ensures it is neutral.

Many devices with switches (such as light fixtures and some appliances) are at least somewhat safer if the switch is on the hot side. Reversing polarity puts the switch on the neutral side.

Reversing hot and neutral may not allow safety devices such as GFCI to work properly.

Even if the current homeowner won’t use any devices requiring polarized plugs for safety, that socket may still be in use decades from now, and perhaps by families. It’s not a difficult fix and should be done. I was shocked more than once as a kid before polarized plugs were required, so I speak from experience.

[RJVB]

I just went through the house: looks like the reversal was indeed made at the breaker, but only on that particular one. Serving sockets, that is. I haven't bother to test the ones dedicated for lights or heating.
I think I'll wait for the next daytime power outage to have a look if I can fix this easily enough myself.

Lightbulb sockets are hazardous if the threaded part of the socket is on the hot side of the circuit. The polarized plug ensures it is neutral.

Neutral still doesn't guarantee 0V AFAIK, and I must add that I've never seen a lightbulb socket where either of the hot parts are exposed to the world. The outside is always an isolator and even the thread is usually plastic. You'd have to unscrew the bulb and stick a finger in (i.e. be very stupid) to get a shock.
(We also have old-style bajonet sockets and those are probably a bit less safe though I don't have one at hand to check.)

The interesting question is how the ground differs in grounded and ungrounded AC/DC PSUs. Would it even be possible to connect the DC neutral to the actual ground - I guess not, right? Yet I noticed something with the variable voltage PSU I bought to drive a mini amp. It's got a pot to vary the voltage that also has an on/off switch. If I turn it off there I get that tingling sensation when I drag a finger over the amp, the kind I also get with my Macbook Pro when using the PSU without the grounded cable. I haven't yet checked if that also happens on a properly wired socket.

Fun fact : the socket for an electric rasor that you'll typically find right next to a bathroom sink does not have to have ground here.

[va3ux]

Exact, and there are strict rules about which devices should have a grounded connection and which can do without. Interestingly (for someone used to using a traditional central heating radiator as a quick "ground source"), our big electric heaters are not required to have a ground connection.



I don't know how residential 110V is produced in the US but here in Europe the (traditional) 220V (now rather 230-240V) comes from what we call tri-phasé, where local power stations would divide a single 380V line into 3 220V lines. Exactly how the current alternates between the phase and neutral poles in one of those lines would depend on the power draw in the other 2 lines, meaning that "neutral" doesn't always mean 0V.

It's actually quite likely that I replaced/installed the actual plug myself; we had to keep the electrician bill as low as possible.

It's done slightly differently in EU. 380V 3 phase produces 220V Line to neutral, and the neutral is grounded. So you also have a 'hot' and neutral like we do in North America, but your 'hot' conductor voltage is 220V relative to ground whereas ours is 120V relative to ground (in the case of residential distribution). Your 'hot' conductor is actually one of the 3 phase conductors in the 380V distribution system. In North America, 120/240V is derived from a pole mounted transformer that is fed by one of the higher voltage 3 phase conductors. The higher voltage in EU and other areas means the current drawn is roughly half of the current we would need to power the same device at 120V. And that means smaller copper wiring and lower resistive losses (for you). Some countries use 400 or 415V 3 phase, which produces 230V or 240V Line to neutral respectively.

[RJVB]

and the neutral is grounded.

Are you really certain about that? It seems to me that if that were true, no separate ground wire would be required, and it also doesn't fit with the notion that the absolute potentials on the 2 poles in each residence depend on the power draw in the other 2 residences. Now it must be about 30y since someone explained that to me, and it was in a different EU country so it's possible that there have been changes.
FWIW, among the first things that were done (to establish a quote for the work required) the different electricians we contacted all checked if the grounding pin in our cellar was up to current requirements.


I think I've also heard it claimed that the lower currents actually make getting shocked less dangerous, which makes sense in a way but also seems improbable. At least when wall outlets are typically limited by a 16A fuse....

[thelostboss]

I think I've also heard it claimed that the lower currents actually make getting shocked less dangerous, which makes sense in a way but also seems improbable. At least when wall outlets are typically limited by a 16A fuse....

From my days as a working electrician; "It's the volts what jolts and the mils what kills" - mils being milliamps. There also appears to be some correlation between electric shocks and later MND according to some sources.

[RJVB]

From my days as a working electrician; "It's the volts what jolts and the mils what kills" - mils being milliamps.

Biologically that makes sense - current is electrons coursing through you, presumably being ripped from their atoms rather than just a bunch going from A to B so even without e.g. the burns you'd still be inflicting structural damage. I'm confident you could kill someone with "just" a 12V car battery, and I still remember the pain when I was stupid enough to grab the 2 wires from a model train transfo in opposing hands, with 1 wire touching a small open wound I had in a finger.

MND being motor neuron disease I suppose?

[rpjazzguitar]

I don't understand the physics of it. From what I gather, you could touch the neutral with one hand and a cold water pipe with the other hand and you might get a shock but you wouldn't die. Meaning neutral isn't necessarily zero volts. Well, maybe I've got that wrong. And, I wouldn't imagine that a modern appliance would connect either neutral or hot to the chassis. Not sure about an older appliance.

Anyway, I may be babbling about that. But, what I do think I understand is that there are reasons that the codes are written the way they are and they were written by experts who probably knew a lot about potential risks. The code is there to protect the users (residents). So, I'd fix it on general principles even if I didn't know exactly what the risk is.

Although it shouldn't happen, I've seen chassis's go 115v hot. Once in a toaster, once in an amp.

[RJVB]

Although it shouldn't happen, I've seen chassis's go 115v hot. Once in a toaster, once in an amp.

In a toaster (or electric heater) it's pretty easy to achieve this kind of thing: they're almost disasters waiting to happen in that sense. We had an old one fail not long ago, and given that it tripped the differential breaker and not the dedicated fuse it's almost certain that one of the wires touched the grounded exterior (I wouldn't know if it could be either wire, and don't really feel like poking around to find out ). Though those differentials are often so fast that they react even to a classic short.

[va3ux]

Are you really certain about that? It seems to me that if that were true, no separate ground wire would be required, and it also doesn't fit with the notion that the absolute potentials on the 2 poles in each residence depend on the power draw in the other 2 residences. Now it must be about 30y since someone explained that to me, and it was in a different EU country so it's possible that there have been changes.
FWIW, among the first things that were done (to establish a quote for the work required) the different electricians we contacted all checked if the grounding pin in our cellar was up to current requirements.


I think I've also heard it claimed that the lower currents actually make getting shocked less dangerous, which makes sense in a way but also seems improbable. At least when wall outlets are typically limited by a 16A fuse....

A couple of very common misconceptions here : the neutral wire is a current carrying conductor (it's the 2nd half of the circuit), it is grounded ( for residential and light commercial use), but it is not a grounding conductor. A ground wire (grounding conductor) is not a current carrying conductor; it's purpose is to, 1) maintain exposed metal parts at ground potential so that there is no voltage difference between exposed metal parts and a human touching those parts, and 2) provide a short circuit path back to the source in the event of a ground fault so that the breaker trips (or fuse clears) thus removing the shock hazard. The ground wire is a 'safety wire' that normally has no voltage or current on it. That's how it is here in NAM and in many other countries. However there are several grounding 'schemes' in use around the world, some of which would not be allowed here in NAM.

Regarding shock potential, the higher the voltage, the higher the risk of receiving a dangerous or fatal shock. Higher voltage causes higher current to flow through a human body (or any resistor) and current flow is what does the damage. A measly 10ma is enough to create pain; 100 ma is very painful and above 100 ma ( 1/10th of an amp) is electrocution territory. It varies from person to person. 100 ma from a 50 volt source is just as bad as 100 ma from a 240 volt source. A higher voltage will always promote a higher current flow through a person.

[sgosnell]

I've survived 30,000 volts multiple times. The high voltage lead of an old color TV carried that, but the current was very low. It will certainly get your attention, but it's not fatal. I worked in my father's TV repair shop when I was in high school and college, and it wasn't at all unusual to get a bite from a CRT lead. You learned to be careful, but feces occur. The most injury was usually the result of my hand jumping away and hitting whatever was close inside the cabinet, usually sharp metal or the point of a tube. I recall once we were working on a customer's TV, with the chassis out and the lead disconnected from the CRT. The customer insisted on watching closely, and eventually asked "What is that?" and pointed his finger at the lead. The voltage leaped the gap to his finger, he jumped, and left the shop and didn't return until we called him to let him know that his TV was fixed and ready to go.

[RJVB]

A couple of very common misconceptions here : the neutral wire is a current carrying conductor (it's the 2nd half of the circuit), it is grounded ( for residential and light commercial use), but it is not a grounding conductor.

I guess the question is where. As I said, I'm not about to test it here, but I'm expecting at least one protection to trip when I connect the neutral to the ground, certainly when the circuit is closed.

2) provide a short circuit path back to the source in the event of a ground fault so that the breaker trips (or fuse clears) thus removing the shock hazard.

The way I see it it's a very low resistance path to a safe "sink". I've never tried to understand how a differential breaker works but I wouldn't be surprised if it is connected to the same ground wire.

The ground wire is a 'safety wire' that normally has no voltage or current on it.

Here it's connected to a copper rod that goes a certain distance into the earth, somewhere (typically the basement).

100 ma from a 50 volt source is just as bad as 100 ma from a 240 volt source. A higher voltage will always promote a higher current flow through a person.

It will of course, but in practice electricity sources are all designed to handle a certain wattage so max current is typically inversely related to voltage. That's why the shock from the high-voltage CRT lead just hurts and probably why my uncle warned me to be very careful fooling around with the huge capacitors in the power unit of an old powerful bulb flash.

Breakers for regular wall outlets are normalised to 16A here, curiously above the 3200W limit you see on many extension cords and stuff. I'm guessing that 110V breakers for the same application will be set to about 32A or some value close to that?

[nevershouldhavesoldit]

Breakers for regular wall outlets are normalised to 16A here, curiously above the 3200W limit you see on many extension cords and stuff. I'm guessing that 110V breakers for the same application will be set to about 32A or some value close to that?

Standard residential branch circuits in the US are 15 or 20 amps. New or recent construction uses 20A, but 15A circuits are still common in older homes and apartments.

[sgosnell]

There are some 30 amp circuits, but they're not common, AFAIK used mainly for specialized applications, and require heavier wire than what is used in 20A circuits.