The obvious follow up question, then, is: why does Japan use 2-prong outlets even today? Are Japanese people dying from electrical shock at a much higher rate than the rest of the world? Is Japan full of only Perfect Electricians? Or is this not actually the significant problem that this myth (in the etiological sense, not the fictional sense) would lead us to believe?
The only 2-prong device I own is a (Japanese-made, of course) rice cooker. You'd think that if any device warranted 3-prong safety, it'd be a metal container which combines water and mains electricity.
> The obvious follow up question, then, is: why does Japan use 2-prong outlets even today?
Because JP is behind the times:
> Japan also uses a grounded plug similar to the North American NEMA 5-15.[49] However, it is less common than its NEMA 1-15 equivalent. Since 2005, new Japanese homes are required to have class I grounded sockets for connecting domestic appliances. This rule does not apply for sockets not intended to be used for domestic appliances, but it is strongly advised to have class I sockets throughout the home.[52]
In the same way the old 'best practices' in automobiles didn't mandate ABS or air bags, some country's electrical best practices aren't written to the state of the art.
Everyone is probably looking at IEC 60364 for overall guidelines:
The Europlug is only supposed to be used on double-insulated appliances - those with an insulating outer shell enclosing all live components. Appliances like that do not require a protective earth as a human is not able to contact the live circuitry under normal circumstances. Same reason as light bulbs are not earthed.
- Japan runs at 100V, the lowest and therefore safest voltage in the world. At least when it comes to electrocution. I suppose that it is also the reason why in Europe, where the standard voltage goes up to 240V, electrical safety is much more strict.
- As it is already mentioned, Japan is backwards in several domains and that may just be one of these. That's despite their technological advance in others.
- I have a feeling that there are more "perfect electricians" in Japan than anywhere else, because of their culture.
- Some 2-prong devices are perfectly safe because they are double insulated. This is also called Class II and the symbol for it is a square inside another square.
Japan runs at 100V, the lowest and therefore safest voltage in the world.
100V, 240V, if 15 amps runs through you, you’re having a really bad day either way. That’s another way of saying, “it ain’t the volts, it’s the amperage.”
So given relatively constant resistance, voltage is linearly proportional to amperage. Double the volts, double the amps.
The resistance of the human body can vary, for instance, if you're sweating, your resistance drops, therefore the amperage increases. Dry skin has high resistance, so if you stab electrodes through your skin and into your flesh, the low resistance of your mostly salt water body will carry a high amperage. But with all environmental factors being equal, if one person gets zapped by 100V and another person gets zapped by 240V, the person zapped by 240V will have 2.4x the amps running through them, and their day will be roughly 2.4x worse.
In other words, sure, guns don't kill people, bullets kill people. But you'll find that whenever a bullet kills sometime, there was probably a gun involved.
I was always taught “volts jolt; mills kill”. i.e. high voltage will knock you across the room but a few milliamps of current across the heart will easily kill you.
So much misinformation in this thread. Voltage, current, and resistance are physical properties entirely dependent on each other -- increasing one affects the value of others. That's like saying that the mass of bullets kills people, without any regard for velocity.
A 12 V car battery can kill you, if it has a low-resistance path that goes through your heart. Touching the terminals will do nothing. Maybe if you applied that voltage to conductive spikes inserted into specific points of the body, there would be some effect.
A voltage source, like a car battery, does not determine how much current passes through a circuit. That is determined by the circuit's resistance, and the source's upper limits (i.e. internal resistance for batteries, rated current for power supplies).
As voltage increases, so does current. High-voltage supplies will absolutely kill you, as long as they're able to supply the required current. Which, to harm a human heart, is quite low (tens to hundreds of milliamps). Aside from fringe sources like static electricity and tasers, most HV sources are entirely capable.
But voltage is what matters to pass through the skin. 12v can kill if it's applied below the skin into the highly conductive inside of the body. But to get through skin you need at least 30v.
I recall a story, late 1990's to early oughts, of a solo researcher found dead in his lab of apparent heart failure whilst working with very low voltage- and amperage currents, but at critical frequencies. I cannot find an online reference presently.
Amps are a side effect of the thing that is killing you - volts.
The only time a voltage sufficient enough to kill you, doest kill you, is when it is only present for an extremely small amount of time and/or not traversing your heart.
> The only time a voltage sufficient enough to kill you, doest kill you, is when it is only present for an extremely small amount of time and/or not traversing your heart.
The reason this is the case is because your initial premise is incorrect. Power kills you. Power equals work divided by time. This is why huge voltages over short periods of time don’t kill you.
> Some 2-prong devices are perfectly safe because they are double insulated. This is also called Class II and the symbol for it is a square inside another square.
Interesting. FWIW, besides the UL logo, my rice cooker has no safety or regulatory markings, and the Zojirushi webpage makes no mention of double insulation or any extra safety features.
Japan runs at 100V, the lowest and therefore safest voltage in the world.
That's not really how electricity works is it?
My understanding is that with the resistance of the human body, once you exceed a certain voltage AC (much lower than 100 V), yes, current increases, but you're going to be severely electrocuted regardless.
Voltage does matter. If you have thick skin or touch cable through something (like layer of paint) it might be enough to save you if the voltage is 100V but not if it's 240V.
ElectroBoom has a great video on how the interplay between current and voltage works with lethality in the human body. As he shows, even 150A in your mouth can be safe:
This is entirely "stuff I absorbed growing up", but I was very firmly given the impression that I shouldn't mess with the 240V outlets, because the 120V outlets might kill me but the 240V outlets will kill me (of course, in the US, the 120V outlets are 15+/-5A, and the 240V outlets are 30A).
How does this work? Most of the places I lived in in the 1980s and 1990s are still standing (incl. houses, small apartment complexes (say 60 units total on 3 stories) and high-density apartment towers)
I was mostly in Tokyo NW, Saitama and Nagoya (and a little bit in a more Osaka-ey direction from time-to-time).
Short form: Neutral is grounded. The "mouth" part of the plug is the ground and is isolated from the neutral to prevent inadvertent carrying of potential.
Aside: it is generally considered safer to have the outlets ("receptacles") installed with the ground at the top, but we don't tend to do this in NA because ground at the bottom makes it look like a little face and humans like faces.
> Aside: it is generally considered safer to have the outlets ("receptacles") installed with the ground at the top, but we don't tend to do this in NA because ground at the bottom makes it look like a little face and humans like faces.
I'm not sure there's any basis for this. See this article for example:
The real reason why this was done historically was perhaps due to this:
> Back in those days, there were very few plastic faceplates on AC receptacles and switches. They were almost all made of metal. All that holds a faceplate in position is one short screw for an AC receptacle. If the screw loosens and falls out, the faceplate will drop down and short out the two power prongs on anything that is plugged into that outlet. Such accidents used to be commonplace which is why power receptacles were put in upside down.
With plastic faceplates it no longer applies. Which is why most newer medical facilities have them in regularly facing. It was done for a reason, but people forgot what that reason is, then myth turned into legend and people kept doing it cargo cult style without knowing why it started ("safer" requires an explanation). The whole "drop a metal tool" justification was likely a rationalization to a practice people didn't understand, but doesn't actually make sense.
That all being said, it isn't less safe to put them "upside down." It just might interfere with some consumer electronics that assume a certain orientation (like those plugged in perfume bottles, smart home sensors, or similar).
This is exactly correct, and electrical code still requires the "faces" to be upside-down if the electrical box has a metal faceplate, which is still common in many industrial settings and also house basements. If the faceplate is plastic, it doesn't matter.
Also, in many houses you'll see the outlets oriented both ways in living and bedrooms, and there's a reason for this: what you'll see is all the outlets oriented one way (usually with the "face" right-side up), except for one outlet, which is upside-down. That outlet is the one that's controlled by a wall switch, so you're supposed to plug a floor lamp into it, which can then be switched on with the wall switch when you enter the room. Inverting that one outlet makes it much easier to see which outlet you're supposed to use.
This is true, but I really hate this standard. They should use a slightly different face-plate color for both outlet and switch to make it even more obvious.
They can't do that: that would cost a dollar or two extra and would require buying different parts instead of using the exact same part for every outlet/switch in the house.
From what I've heard (UK), earth at the top is marginally safer as if anything falls on a partially unplugged plug (coin, hairpin, something metal), it will land on the earth rather than directly on the powered pins. Of course, this is also resolved in the UK by insulating the lower portions of the powered pins, so whenever it is powered only insulated area will be visible outside of the socket, so it's something of a moot point.
That all sounds plausible, and I just don't remember where I've heard that "ground up is safer", but I'm also possibly colored by having experienced something falling between a wall wart and the receptacle when I was younger and that left an impression. Ground up wouldn't have helped in this case since the wart was a 2 prong, but it seems like there are opportunities where it could help and few where it hurts...
I'm also just a home gamer, though I have done my research via reading relevant parts of the code for wiring and passing inspection on my kitchen remodel including adding circuits, pigtailing my aluminum wired house, etc... It was somewhere in there I came across the "safer, it's code in hospitals" idea. Though my memory is that the hospitals I've seen all had metal faceplates, but that's just from memory.
I see you've not dropped any kind of conductive flat sheet off a desk and have it land between the plug and socket when installed in the smiley face orientation. Sparks and fire isnt fun.
It requires the flat conductive sheet to somehow re-orientate itself in the air from horizontal to completely vertical in the two feet between your desk and the top of the plug (unlikely).
If this is a problem you regularly encounter, this hack isn't the safe solution, the safe solution is an Recessed outlet/Outlet Cover. That way the metal sheets that assault your outlets cannot hit any of the pins, and you aren't relying on luck for how it will fall/rest after it contacts the ground pin.
Most people aren't suffering this issue, so a solution isn't needed.
It is all about risk mitigation. When building for long term use and safety, scenarios like a short from something falling are designed away.
Think super cheap insurance. And liability management. People can perform to code and that compartmentalizes damages and responsibilities. Couple that with licenses and it's all public safety oriented.
All of it adds up.
Where risk is higher, say in the parent example, yes. By all means use a product intended to manage those risks.
Costs are marginal otherwise, might as well take the cheap insurance.
> It is all about risk mitigation. When building for long term use and safety, scenarios like a short from something falling are designed away.
Then you'd want to install Recessed outlets, Outlet Covers, or GFCI Outlets. Since they're designed to be safer, this simply relies on luck that the conductive object bounces off of the neutral pin and away (which may not occur).
Plus the cost isn't free as has been discussed in this thread elsewhere. It breaks a lot of consumer electronics.
GFCI won't protect against a conductor shorting the neutral and hot. It WOULD trip if the outlet were installed with the ground up,and a flexible conductor, say a necklace, fell between the plug and the receptacle.
Yes, recessed or covered receptacles would be a better option, for some values of better. I mean, most people wouldn't accept covered receptacles in their living space because they are "ugly" and would prevent many pieces of furniture from being placed in front of the receptacles (which by code are spaced every 12 feet at the most). And recessed receptacles are going to be a tight fit in most existing residential boxes. Forget smart devices fitting recessed, it's hard enough to fit those into regular boxes once you add some wire nuts back there...
So, yes, there may be some more effective options, but installing devices ground-up reduces a not uncommon hazard without resorting to changing out boxes or adding covers to everything.
Code makes the minimum calculation - as to how it originally came about I would guess a mix of empirical (committees working with various sized boxes as they work on NEC revisions) and then math to have a formula/table for the real world.
There are wire fill factors in the NEC - basically if you have a junction of a conductor of this size (gauge) then it must have this many cubic inches to go with it.
The simplest outlet (a receptacle) would have 2 conductors (hot and neutral) and those each require a certain number of cubic inches based on their size (12AWG and 14AWG being the most common). The cubic inches of a box are typically stamped on the box and the amount of conductors ('wire fill') must be below the box capacity. Ground doesn't (didn't?) count in the calculation as it shouldn't carry current normally so it's not a conductor.
If you've worked in a box at or near capacity (especially with a AFCI or GFCI/RCD) you will always put in a larger box given the chance.
The likelihood of a conductive sheet falling off a desk on the wall side and reorienting itself to be thin edge down is pretty much 100%, considering that the space between the wall and the desk probably turned it and likely kept it in that orientation when it fell.
It's not that this happens to any one person a lot, it's that with billions of receptacles in use throughout just our country alone (tens of billions? Just a WAG), it collectively happens kind of a lot. I've only experienced it once in my life.
I recently remodeled a house (in the US), and installed every receptacle with the ground at the top, because of that.
Every person who comes over that knows anything about electrical stuff instantly says 'your plugs are upside down'...and then I have to explain to them how/why it works :)
The inspector had no issue with it though, thankfully! In health/saftey critical installs (like hospitals), the ground MUST be on top, according to our local code. It's optional for residential.
Your explanation was likely wrong then, because there's no good reason to install your receptacles that way, unless your faceplates are metal. If they're plastic, then there's no reason for it.
Huh? I thought the reason for putting ground facing up was to prevent metal things from falling in from above and shorting the live and neutral together. What does that have to do with the material of the faceplate?
The reason for putting it ground up is that the vast majority of plug insertions occurs from above, with the plug angled down into the socket. This assures that the ground connector goes in first, as it's at the top, so you get the safety ground before any of the connections are made.
It's also a nice benefit that if something were to fall between the plug and the faceplate that it'd hit the ground and not short things (or short to the plate itself if it were metal).
This doesn't seem right to me. Given how the contacts are positioned deep in the socket, they wouldn't be reached if the plug isn't being inserted straight. Plus the ground pin is already slightly longer to mitigate this issue. Do you have a source for this explanation?
They are deep enough that it's not possible to make contact with the live and neutral pins before the ground pin. Try it, I just did and I wasn't able to do it regardless of angle.
Because a metal face-plate is the most likely item to fall in there. Especially in industrial places like warehouses and hospitals where there's a lot of stuff rolling around and everything is high traffic.
Really I've found flat materials falling off the back ofdesks to be a pretty common method of shorting said prongs. Even upside down it is still a problem with ungrounded devices like phone chargers.
I remember seeing a couple places with their outlets installed this way several years ago and seeking out the reasoning at that time. I never liked the way wall warts fall out of "upside down" sockets, though, so when I had to install a bunch of outlets recently I put them all on their sides with the neutral leg "up".
That's not a bad idea either, although I'm sure I'd get even more questions as to why all my outlets were on their sides!
Also those 1-gang boxes are more expensive...
Nothing so far.
Most all appliances these days are single-device circuits (in the new kitchen, literally every appliance required a dedicated circuit - fridge, gas range, microwave, ovens, dishwasher, garbage disposal, etc.). So, lots of single-plug receptacles (code says a dedicated circuit has to have a special single plug receptacle here). No issues so far with anything.
My brother in law lives in a fairly recently constructed house, and I commented that one of his receptacles was "upside down". He explained that the electrician did that for the switched outlets so you can tell by looking at them. He then asked if there was anything wrong with it? "Nope, it's actually code in most hospitals to do that, but residential doesn't. Just wondered." It's an interesting way of indicating switched.
Usually I find that wall warts falling off is due to the receptacle being "past it's prime" and replacing it takes care of it. I can't think of when I've seen a polarized wall wart, so you can sell place one in the lower outlet facing down, and the upper one has to face up usually because they hang down and interfere with the lower port.
Aside: One regret I have in my kitchen remodel is not replacing all the boxes with dual gang.
I've rarely seen a DC transformer that is also polarized. It's literally a coil of wire passing from hot to neutral. Which way doesn't matter as it's rectified from AC pos/neg to DC +wire -wire.
To extend this short form a little bit: at least for me, a key insight of this article was: Depending on the height of your circuitry above the earth's surface, depending on the area covered by your wires, and depending on whether there was a thunderstorm above you at the time, there might be a fairly huge DC charge on your electrical distribution system.
So, the article offers an explanation of why it's really useful to ground one of the two AC wires in the first place: AC generated by the power company may not be the only source of electrical charge on the wires. I must admit, that was not something I thought about before.
> There is a simple solution to these problems: connect your system to the Earth. Drive some long metal rods into the dirt, and connect them to your wires. That way, lightning currents will be directed into the Earth rather than spreading throughout your power lines.
I have tried and failed many times to understand the concept of "ground." Why does electricity "want to go to the ground?" What constitutes "ground?" Can I touch wire to concrete, and that counts as "ground?" No? Must I then sink it? How deep? Until it touches dirt? What about sand? Where does the electricity "go" then?
I can answer a narrow part of your inquiry: Why do you need to sink the rod into the dirt, and merely touching it isn't good enough?
Dry dirt is a very poor conductor. Wet dirt is pretty good, thanks to the dissolved salts, but still not great compared to copper. Most dirt is somewhere in-between, so the amount by which you must sink your ground rod to be safe partially depends on your local climate.
Lets say you have a 3/16" grounding rod sunk three feet into the dirt. The cross section through the rod at the top of the rod is ~0.025 inch^2. But the total contact area on the local dirt is ~42 inch^2. That's an increase in area by a factor of over 1500x. The local dirt doesn't need to be particularly conductive to work as a reasonable conductor when you have that much contact area to work with.
The total cross sectional area of the dirt-conductor around your ground rod also gets larger quickly, proportional to r^2. That term gets large rapidly with distance, such that almost all of the resistance between the planet as a whole and your ground rod is concentrated at its interface to the dirt.
There are two different mental models I find really useful when thinking about electricity. The second one will answer your question more directly, but the first one helps set the stage better.
Mental Model #1 (useful for most work with electricity):
Think of electricity moving through circuits as water going through pipes.
Electricity = water
Voltage = water pressure
Battery/Voltage Source = water pump
Wires = Water pipes
High/Low resistance = Wide/narrow water pipes
Voltage is like water pressure. You have a pump creating high water pressure is like a power source with a high voltage. The pump is always on and creating water pressure, but there's a cap placed on it that blocks the water from escaping. Similarly, in an open circuit the air acts as this cap and blocks the electricity from proceeding.
The wires in a circuit act as pipes channeling the water to different areas. Resistance in a circuit is analogous to the thickness of the pipes (low resistance == wide pipes, high resistance == thin pipes).
The above analogy sets part of the mental model. The twist is that in reality electrons want to actually travel from a place of low voltage to a place of high voltage (classical electrical diagrams show this process in reverse). Which leads us to...
Mental Model #2
Voltage sources are powerful electron vacuum cleaners trying to to suck up electrons.
(You can apply this on top of mental model #1, just reverse the direction of the water flow)
Open circuits generally don't conduct electricity since within a nano-second the voltage source will suck up all the free electrons in the circuit. The negative/ground pole in circuits are where the free electrons get supplied from for the voltage source to "suck in" the electrons from.
The ground (by virtue of it being so large and having lots of moisture/materials which conduct electricity) happens to have a virtually limitless source of electrons available for the voltage source to suck into itself. This provides the circuit with all the electrons required to successfully operate.
Here's another mental model, the balls model, which I sometimes find useful, especially in this case. Here it is in case it helps someone:
Imagine electricity as red balls, which can travel along conductors. Groups of red balls feel drawn strongly to places which have less balls, and away from places which have more balls, leading them to equalize quantities among all points in the circuit (let's assume for now cables can't store the red balls, only transport them. Points in the circuit are places that can store balls).
if you have a battery, both the sides can store balls, but assuming it's charged, the plus side might have 12 red balls and the minus side none.
The difference in red balls is 12! A quite strong force pushing red balls from the + to the -. (that's the voltage).
When the circuit is closed with a wire, the balls will travel quickly and end up with 6 at both the + and -. How many of them travel at once (Intensity) depends on how large the tube is (Resistance).
What if instead of 12 and 0 we have 112 at +, and a 100 balls at -? Well the force is the same as before (12), the whole circuit works the same way. 6 balls go from + to -, over the same amount of time (tubes have the same throughput), you then end up with 106 at both ends. You just have these 'residual' red balls hanging around, doing nothing special.
However, if you suddenly connect this circuit with 212 balls to the first circuit with 12, the balls will quickly flow from from the first to the second, this time with a force of a 200!
(The point here is that voltage is a relative measure, unlike charge.)
You might have an isolated electronic circuit starting with 12 and 0 balls, and think all is well. You touch the circuit once to start it, the balls are flowing normally. You leave it working for a few hours and come back, it still seems to be fine. You touch it, and ZAP, a spark, burnt smell, you're wondering what happened. Well the thing is that external sources can sometimes add a red ball or two to your system. Over time, they might add up, and after a while you end up with hundreds. As soon as you touch the circuit, they all flow to somewhere that has fewer.
Ok, where does the ground come into this? A 'ground' is like the - side of your battery. It's an object that can store balls the same way. However, instead of having one 'storage point' which can be filled with balls, it has a several. So take a metallic object, and say it has, for example, 11 storage 'points'. If you connect it to a single + point containing 12 electrons, the balls spread out so that there's the same amount of balls in each point (as we explained at the beginning). So after a while your 1 and 11 points, end up with one ball at each point. (Analog to the concept of capacity)
So the natural earth turns out to behave like a huge metallic object. So big that it basically has an infinite amount of storage 'points'. You can pour red balls into it all day, and still end up at an average of zero balls per point.
So now you take your 12 to 0 circuit and connect the - side to the earth. The difference between + and - is the same as before, 12. The flow from + to - is about the same (same resistance), so the circuit behaves almost the same. If however an external red ball is randomly added, it flows from the - to the earth, meaning that you never accumulate balls in the circuit.
This allows you to set a reference amount of 0 balls in every circuit you work with. Which is great. The amount of red balls at the + can still be set to whatever you like, and they will flow nicely from + to - along whatever circuit you create (like before), but on top of that you never run the risk of having red balls jump from one circuit to another if they accidentally touch, because there is never a large difference in red balls as they all have the same reference amount.
Anything big which is conductive enough can store a lot of red balls. So what qualifies? Soil usually works, water works to an extent (https://electronics.stackexchange.com/questions/164898/can-t...). When you can't connect to those a big object will have to do, like a car frame, it will have enough storage that over a short time it looks infinite, but if you really pushed and kept adding red balls at some point the reference level would change. With the earth you could pour red balls your entire life and barely make a difference to the reference level.
(So now you're thinking, the red balls are electrons, right? Well sort of, but don't tell a physicist! Otherwise they'll get angry and start talking about flow direction being this or the other way. Or they'll say something about non-rigorous representations, wave-particle equivalence, the balls actually being a 'lack of electrons' and we'll all be confused.)
Whenever you have two electrical conductors near each other, but not touching, you have a capacitor. Electrical transmission lines are obviously conductive, but so is the earth, with its large amounts of metal and salt water.
When extra electrons collect in the transmission lines[1], they become negatively charged and repel electrons on the surface of the earth, leaving unpaired protons and a positive charge there. As there are two oppositely charged objects in proximity to each other, an attractive force exists between them. This force is generally referred to in terms of voltage because it’s very small in absolute terms, but electrons are also very light and thus strongly affected.
The electrons can’t make it across the gap to the positively-charged Earth, however, because air is an insulator. That means that all of its protons and electrons are bound tightly to each other, so there’s no intermediate place for the electrons in the wire to jump to, and they have to make it all the way across the gap in one go, which is extremely unlikely.
By driving a metal stake into the ground, you’ve provided a path along which the electrons can move little at a time in the direction that the electrical force is pulling them. When they arrive, they’ll pair up with a proton and reduce the attractive force a little bit. Over time, this prevents the large buildups that can become dangerous.
[1] Electrical storms and lightning strikes are the most dramatic causes, but not the only ones.
Whether we're talking electricity or gravity, potential differences are always defined _with respect to something_ -- where you put your zero is pretty arbitrary, and for some problems there's an obvious "right answer". For stuff to do with space, for example, we might want to put the zero of potential at the edge of the universe: at a glance, anything with positive energy therefore has enough to get there (and is "free" from any ground gravitating body) and anything with negative energy is "bound" to never get there (and typically bound into orbit around something else).
It's the same idea for electricity. Potential differences are what drives currents: at some level, that boils down to more electrons over here, more electrons over there.
The thing about the ground is that, usually, we're all on it. From a safety point of view, this matters, because we're slightly conductive bags of meat that rely upon electricity in certain bits to live. Bugger up the electrics, stop us living. As we always have one foot on the ground, the potential of the earth is a good reference point to chose, and it's the convention that for electrical thingywhatists, _it's_ the zero of potential, i.e. 0 V. By definition, therefore, electricity is anything that develops a potential difference with respect to this -- it will generate an electric field to move charges around, and the charges feel a force as a consequence.
Conveniently, unless you live in a desert soil is itself quite conductive, and a long (typically >4 m long) rod of metal driven directly into it. Often your power company will do this for you as the niggly details about "how big", "how many" and "how wet is the soil" turn out to be quite important, and also quite complex. Concrete doesn't easily work as it's an insulator: wet, loamy soil works brilliantly. People have written whole books and PhD theses about how to deal with granite, sand, and similarly frustrating geology.
Ordinarily, in most power distribution systems around the world, the flow of charges -- the electrical current -- doesn't actually use this ground all that much. You can think of it coming in via live and leaving via neutral, if you want. The permanent earth is a protective feature to ensure that a very large current _could_ flow to the ground if it got the opportunity to, and _not_ flow through you. This is a safety device in disguise, as it'd trip a circuit breaker and protect the wires in your house from catching fire.
Mains electricity is a bit different: because of how generators work, it turns out to be easier to generate (pun not intended) three _different_ live wires all at the same time, called L1, L2, L3 or three different phases. They are all AC voltages, developed with respect to each other, 120º out of phase. At the substation that transforms these to lower voltages, the central tap of the transformer is hammered into the ground, and your neutral wire -- which is also connected to that central tap of the transformer -- therefore is about ground too. This provides a nice, handy, readily accessible common reference that both prevents you frying and lets your electric hob continue doing so at the same time.
Incidentally, this is why if you're running an electric fence, and you have rocky or dry soil, it is advisable to run alternating hot & ground wires, and to not rely on the ground to complete the circuit.
> Before you "grounded" your system, the AC voltage in general acted pretty safe for your customers. The only way they could get a shock was if they touched both wires at the same time. This was a fairly rare occurrence. One single wire acted as if it was "safe," and it did not deliver shocks.
I'm confused by the assertion that you could not get a shock from a single wire in an ungrounded system. If there is 120V AC between the two wires, doesn't this mean that there has to be at least 60V AC potential between at least one of the wires and earth ground? Is he depending on the "halved" voltage for safety, or depending on having an intact circuit so the other connected wire is always "better" than an earth ground, or is he right in some way that I'm not seeing?
The statement is predicated on the entire system being ungrounded (i.e., everything on your side of the transformer). Voltage is relative; if there's no prior connection between two systems, when one gets connected to the other, the systems become referenced to each other at that connection, through which no current can flow (since it's the only connection), and hence exhibits zero voltage across it. In this example, the other half of the power outlet is then 120 V relative to whatever the person is touching (presumably Earth ground).
(Also side note: in the US system with hot and neutral, neither is 60 V from ground. Hot is 120 V from ground, neutral is (ideally) 0 V. Only in a 240 V system are the voltages symmetric.)
Note also that the typical residential electric service in the US is actually 240V split-phase; the distribution transformer secondary winding is center-tapped, with the center tap grounded and brought in as the neutral. Phase-to-ground is 120V, phase-to-phase is 240V. Normally, 240V is reserved for heavy-draw appliances like large air conditioners, ovens, arc welders, and so on.
In some apartment buildings, commercial properties, and densely-populated neighborhoods, two phases of a 208V three-phase supply are used instead, which still yields a 120V phase-to-neutral voltage. Large appliances often have a wiring option to operate on 208V.
[ETA: The second hot wire is usually red, and 208V three-phase systems use black, red, and blue for the hot wires.]
> Note also that the typical residential electric service in the US is actually 240V split-phase; the distribution transformer secondary winding is center-tapped, with the center tap grounded and brought in as the neutral.
Why is this the case historically? It's been a huge annoyance for me personally, as I want to use my arc welder in my garage that has only 120v circuits, as opposed to my cluttered shop that has a 240v circuit available.
> I want to use my arc welder in my garage that has only 120v circuits, as opposed to my cluttered shop that has a 240v circuit available.
I hear you on this; I'm in the situation where neither my small shop nor my garage has 220-240V available. Also, I don't know if my current service will allow for another such circuit - in any case, I'd have to likely add an auxiliary panel to accommodate the extra breaker(s) needed, as my main panel is full.
But I do have one possibility - and maybe you do, too?
From the door to my garage that leads into the house, the laundry room is just around the corner. In that room, I have both the dryer and water heater, on separate circuits. So in theory, I have the voltage needed.
Now, the water heater is hard-wired, so nothing I can really do there (well, I could - but I won't). The dryer, on the other hand, is plugged into a 220V outlet (as they usually are).
So - what I have considered doing is making an extension cord; one with a plug for the dryer, and on the other end a socket for the welder. In case you aren't aware, the plug on a welder for 220VAC is different than the plug on a dryer. And technically, you're not supposed to use a welder on such a socket (there's something in the code that forbids it).
But people have been making and using such extension cords for almost the entire time such welders have been available, and I have not heard of any reasons against them. You can even purchase such cords thru some places, so they're even legal to sell, purchase, and own. It may be something having to do with the strain a welder puts on the plug and extension; that the extra connection can heat up too much?
Anyhow, maybe you could do something like this; get an IR thermometer and monitor the temperature of parts as you weld and use the circuit; see if there is any issues in that regard.
Another thing I've thought about doing, since the wall where the socket for the dryer is shared with the garage, is mounting a junction box, bringing in the 220, then tapping off it inside the junction box to a new socket and the old dryer socket. Then I could plug right into the socket in the garage without needing a custom "extension cord", so long as I unplug the dryer (maybe add a cover to both as well that can be locked to prevent both being used at the same time).
Something tells me, though, that such a setup would not be up to code - but I don't know for certain.
In North America, The different shapes of plug for the same voltage represent different current-carrying capacities for the wires they’re attached to. There are twist-lock and non-locking versions of each.
If you’re using an adapter to plug a high-current device into a low-current socket, there is a risk of fire from putting too much current through the in-wall wiring. Tripping a circuit breaker is more likely, though.
If the plugs differ and both are the straight blade type, it's because something relevant to safety differs, probably which wires are present (white neutral or green safety ground) and/or the current rating.
You can get in trouble if the welder expects a safety ground or neutral intended to carry current that you don't have (neither is safe substitute for the other). You can get in trouble if you connect the welder to a circuit that can deliver more current than it's designed for and something fails (it was designed to fail safely assuming that the circuit is limited to X amps by a fuse or breaker).
If the dryer outlet has all the wires the welder wants (and some extra) and the same current rating, an adapter that connects the desired wires and ignore the rest should be safe as long as it's made from big enough wire.
If you are doing this, make sure to use a double pole circuit breaker on the lower current side to make sure that any electrical faults are interrupted correctly.
> The statement is predicated on the entire system being ungrounded (i.e., everything on your side of the transformer).
This is not possible for AC due to capacitive coupling between everything. In a large-ish ungrounded AC distribution system, a fault from one phase to ground will carry a large amount of current due to capacitive coupling from the other phases to ground. To reduce this current, an inductor can be installed from neutral to ground [0].
Even for use within a single room, a safe isolated AC system is complicated [1]. If you’re doing benchtop AC experiments, you should not use an isolation transformer — you should use a GFCI.
In fact, you can buy transformers that are ungrounded to isolate the supply and make it safe to touch either wire. When combined with an RCD, this makes a very safe system
In the UK, isolated transformers are how we allow shaver sockets in bathrooms. You touch one of the wires and (via wet feet a pipes) you create the path to ground, which is bad. But because the wire is only live when there's a complete circuit, it's safe. You'd have to grasp both sides of the circuit with different hands to cause major problems
A bonus for the shaver socket is that it is often switchable, so a 120V shaver can be plugged in. This is less of an issue now, with newer shavers having circuitry to allow them to run on anything from 100-240V.
But the ungrounded transformer prevents the fault currents that the RCD would detect, so unless I am mistaken there is no scenario where a RCD would provide additional protection beyond an ungrounded transformer.
Are you saying that the DC voltage difference is always 0 VDC? Since much of the article is about the necessity of grounding your electric distribution network to prevent high static voltages, what makes this system different? Instead, I think it's more likely that the difference in DC voltage can vary significantly.
Are you saying that the AC voltage at all points in the isolated system is always 0 with reference to earth ground? Again, this seems unlikely. If you take 3 simultaneous measurements (Isolated A to earth ground, Isolated B to earth ground, and Isolated A to Isolated B), I have trouble believing that you will find 120V between A and B and 0 between both and ground.
So rather than there being "zero voltage", I think the reason for safety is related to dfox's answer below: the system is limited in how much current it can put through you, and it's the amperage that harms you rather than the voltage. You might get an initial static shock, but the AC amperage is going to be small enough not to harm you. But I'm still confused why this is the case. I think it may have more to do with fusing and generation power than with isolation per se.
Yes on the side note. The article does a great job of explaining (at least for the US) why neutral and ground are distinct, and why hot and neutral are not symmetric.
First, a side note: when you, as someone unfamiliar with the subject matter, say things like "I have trouble believing" and "this seems unlikely" to someone with subject knowledge (I have an EE degree), you come across as condescending. "I don't understand" is a better way to phrase your confusion than doubting the veracity of the person you're asking for help from, and makes them and others more likely to continue to help you.
> Are you saying that the DC voltage difference is always 0 VDC?
AC vs. DC makes no difference in the statement I made. However note also that DC is not transferred through transformers; any DC offset with respect to earth ground in the distribution system will not be seen by the customer.
> Are you saying that the AC voltage at all points in the isolated system is always 0 with reference to earth ground?
No. But if you measure one at a time, with no other connection between the systems, you will see this. If you measure two simultaneously, what you see will depend on the relative impedance of the two measurement devices. This is easily experimentally verifiable.
> So rather than there being "zero voltage", I think the reason for safety is related to dfox's answer below: the system is limited in how much current it can put through you, and it's the amperage that harms you rather than the voltage.
No. Those statements are not separable. See andyjpb's sibling comment.
Sorry, this wasn't my intent. From your other answers over the years, I recognize your username and actually have a lot of respect for your expertise. I don't doubt your veracity at all, I was just trying to make sure you were focused on the question I was interested in, and encouraging you to go deeper in your explanation of why the (to me) counterintuitive answer might be true. Your side note in the original about the difference between neutral and hot wires made me think you might be answering a different question that the article already covered well, and that I feel I understand.
> However note also that DC is not transferred through transformers; any DC offset with respect to earth ground in the distribution system will not be seen by the customer.
Yes. I was talking solely about the measured DC on the isolated side. This is why I think we might be talking about different question, or at least at aiming at different levels.
Maybe (if you'd be willing to continue) I can substitute a related question that I think addresses my misunderstanding: how is an isolated transformer (powered by an AC current) different in safety considerations than a small electrical generator (powered by spinning magnets)? In my possibly mistaken mental model, a small generator is "safe" in the same way that small transformer would be, but as it grows larger the potential for hazard increases. Is this intuition wrong?
> Yes. I was talking solely about the measured DC on the isolated side.
Ah sorry my mistake. I thought you meant "distribution system" as in "the grid".
Yes, safety-wise, grounding (as opposed to isolation) protects from shocks due to DC static charge, precisely by providing a reference point so that the isolated system cannot "float away". However note that that DC charge is temporary: as soon as something gets "shocked" by it, it is discharged. It is still a safety concern, but of a different nature than getting a continuous shock from AC.
(To further clarify: my initial response to your statement about "60 V" only pertains to the AC voltage intentionally present in a system.)
Additionally (I think the article gets into this, but it is long and I am short on time), isolation is not sufficient for household safety because it only protects against a single accidental connection. As soon as a second, different, point on the AC system makes an accidental connection, you have a circuit and (potentially lethal) current will flow. And, given the number of appliances in modern homes, it's not unlikely that one of them might have an accidental connection to earth ground. If a human then makes a second connection to the other leg of the circuit via some other appliance, game over.
The modern grounding system solves this by forcing all appliances to be designed in such a way that either it's "impossible" for a person to contact either leg of the circuit, or that the entire casing of the appliance is connected to earth ground, so that if the case does get connected to a live wire, the circuit breaker gets tripped. (However I'm approaching the limits of my knowledge here – I have no training as an electrician – so won't make any claims as to whether this is the primary function of the grounding system, vs. preventing static charge.)
BTW if you're interested in this stuff, any material you find online by Mike Holt is very good. He is an unofficial authority on all things US electrical code.
Would refer to NEC 250.21 for discussion of "ground detector" as it relates to the detection accidental first ground connections in ungrounded systems.
When you say things like “you come across as condescending”, the author can feel hurt, because the story in their mind is that a friction in this one interaction is seen as an inherent flaw of the author or of their writing. Hurting your interlocutor’s feelings can actually further sabotage constructive communication, if that is your goal.
I’ve found more success by communicating my own feelings as directly as possible, without theorizing. For example, “When you asked me a question about my area of expertise, and then dismissed my advice, I felt disrespected, because it makes me think that you never sincerely wanted my help in the first place. Do you?”
While I'd strongly prefer that colanderman not be offended, given that he was offended, I'm glad that he voiced his offense as bluntly as he did. Yes, it hurts to think that my writing is that unclear, but I've got to own the audience's reaction even if (especially if) it's not the reaction I was aiming for.
There is no way to have a fully isolated system. At best you get either a static charge or at worst, multiple the voltage and same current flowing to actual Earth due to capacitance. As in literal ground.
Most isolators end at some megaohms in a range of temperatures and non-negligible capacitance. (They're used as capacitor fill for that reason.)
Useful personal isolation and grounding is meant to strike the balance where you still get trickle grounded while not being fully isolated and accumulating dangerous charge.
It might be more intuitive if you think of the wires as the two sides of a battery. You can safely touch either the + or - side of a battery, even a high voltage one.
When you have a power source it needs two wires to make a current flow: a wire "out" and a wire "back". You're only part of the circuit if you're connected to both sides.
When the two wires are not connected to anything, there is infinite resistance between them.
If you grab one side then there's no path through you to the other side so no current flows: the resistance between the two wires is still infinite.
Due to Ohm's law, (V=IR), if the current through you is zero then the voltage across you is also zero. The hand touching the wire is at the same voltage as the rest of your body.
So you "float up" to whatever voltage the wire you are touching is at.
If you complete the circuit then current starts to flow. The shock you receive is related to the current that flows through you. However, even tiny currents cause a big shock because humans are quite sensitive to currents in the micro or milliamps range as that's more than our nervous system uses to move our muscles.
> When you have a power source it needs two wires to make a current flow: a wire "out" and a wire "back". You're only part of the circuit if you're connected to both sides.
Wait, does this apply to AC too? I always believed AC can create current even if the loop isn't closed. Say you have an exposed AC wire and touch it; I'd expect you to get a burn at the place you're touching, as the electrons from your finger at the contact point are pulled into the wire and pushed back to the finger, at 50 or 60 Hz. In general, isn't this how RF burns work?
However, RF is more complicated. The difference between "RF" and "AC" is a matter of what materials and frequencies are involved and how you are using them.
Terrestrial RF systems rely on the Earth as a ground conductor. It becomes more complicated because the impedance (like an AC version of resistance <waves hands>) of a material (such as your body between the antenna and ground) will change depending on frequency. Once a circuit it in place, the impedance mismatch between your finger and the thing it is touching will cause energy to build up at that interface.
This is similar to how light diffracts when it moves from one material to another and you can see the boundary because some energy builds up there and is scattered.
You won't get a burn if there is no circuit.
...but if you are fully enclosed in an RF field, you may get little circuits forming because your body is in contact with that field, and therefore referenced to it, and the potential of the field might vary across your body.
RF creates an electrical field across your body, which then causes a current inside you which makes heat. At 50/60Hz the wavelength is planet-sized, so that effect is nil.
Your body also has a little bit of effective grounding just by existing inside a room, but that's modeled as "a 100pF capacitor in series with a 1.5kΩ resistor". Charging a 100pF capacitor to 120 volts 120 times a second comes out to 0.2 milliamps, which is not going to burn anything. 0.6 milliamps on a 240 volt 50Hz supply. And that's only if your current source is badly isolated. The better it's kept away from the local ground, the less flow you'll see.
Thank you, and I think I agree with everything you say. Yes, at steady state, as long as you don't complete a circuit, no current flows. Helicopters can be used to do live repairs on high voltage power lines, and if the pilot keeps the copter from running into anything, no one dies: https://www.youtube.com/watch?v=IgrN0fyDRsQ
Maybe what I've failed to make clear is that my real question is about the relative difference in safety of the isolated transformer as used in a "razors only" plug and the normal breaker protected outlets used in US houses. For both of them, you could stand on a dry wooden ladder and momentarily grab either wire. But the focuses on the increased safety of the isolated transformer. I'm trying to understand what exactly about this system provides for the increased safety.
From what I can tell, everything in your answer applies to both systems. It's the difference in safety between the systems that I don't feel I understand.
"Normal breaker connected outlets used in US houses" are already connected to ground on one side. If you grab the neutral side, you will (ideally) get no shock. If you grab the phase side then you'll complete the path to the other side, through ground.
An isolating transformer is a 1:1 transformer where the input can be referenced one way and the output another.
The output is not connected to ground on either side so when you grab one of the sides, that's the first time it gets referenced to ground. ...and then the explanation continues as per my previous post.
There are a large number of serious errors in this comment which I will try to address.
> Are you saying that the DC voltage difference is always 0 VDC?
Measured across the only connection between two otherwise isolated systems, yes.
> Since much of the article is about the necessity of grounding your electric distribution network to prevent high static voltages, what makes this system different?
The size of the system makes a big difference. The power grid is large enough that you have to worry about things current induced by geomagnetic fields, lightning, solar activity, etc. Lightning has a high enough voltage that it breaks down insulators that would otherwise isolate two systems.
Basically, no system is perfectly isolated from another. It is a question of how much leakage current you have compared to how much current you are using to get work done. With ordinary home wiring, leakage current is negligible, because the resistance of the leakage current paths (many MΩ) is huge compared to the equivalent resistance of the circuit (maybe 100Ω).
Your home wiring system cannot withstand a lightning bolt, it would destroy much of the electronics in your house. The power grid must withstand lightning bolts on a regular bases.
> Instead, I think it's more likely that the difference in DC voltage can vary significantly.
Why do you think this?
There are real electricians and engineers on HN, people with domain knowledge who aren’t guessing but have actually designed power supplies or fixed faulty home wiring systems.
> So rather than there being "zero voltage", I think the reason for safety is related to dfox's answer below: the system is limited in how much current it can put through you, and it's the amperage that harms you rather than the voltage. You might get an initial static shock, but the AC amperage is going to be small enough not to harm you.
Some errors:
- House wiring is not limited in how much current it can put through you. It will kill you dead, given the opportunity. The fuse box or circuit breakers won’t prevent you from being electrocuted, they are there to prevent fires.
- Phrases like “it’s the amperage that harms you rather than the voltage” are a bit simplified. The relationship between voltage and current is fixed. Recall Ohm’s law for resistors, you cannot have current without voltage (except in a superconductor).
The reason why touching one wire is “safe” is because the return path for the current is usually very high resistance—it goes through your shoes to the floor, and through the floor back into the circuit somewhere. This is usually a dubious path for current. But there are a lot of reasons why there might be a better path for current—maybe you are touching something made of metal, maybe the floor is wet and you are not wearing shoes, maybe…
This is why one wire was chosen to be “neutral”, so it would be “safe” to touch even if there was a good path for current. But it turns out that this safety disappears as soon as there is a wiring problem in the house, or whenever there is significant ground current, or one of several other dangerous failure modes, so this is why we have the third ground pin.
I misunderstood that part during my first read, so I would like to add a few clarifications:
> > Are you saying that the DC voltage difference is always 0 VDC?
> Measured across the only connection between two otherwise isolated systems, yes.
The systems are isolated, so there is no fixed potential between them, their "DC" potential difference is therefore random (unknown quantity of excess/missing electrons) before they are connected, and you can have charge transfer if you connect both, that will balance the extra electrons on both sides. Electrons will flow in a way that the resulting density is uniform, like repulsing magnets. Therefore, there is no potential difference between any point in these two systems (barring local differences -- like a battery--, or at the very least, no potential difference across the connecting cable).
An analogy that could or could not help is trough gravitational potential energy: take two water buckets, isolated. The potential across them is the height difference between their water lines. We didn't specify where we left the buckets, so it is random. Now if we connect the bottom of the buckets, the system will find its way to the "ground state", or lowest energy state: the water line in both buckets will align (Communicating vessels), and you therefore have no DC potential across those: connect an extra piece of tubing, nothing will flow.
This analogy is quite powerful and can be used for capacitors, diodes, etc. (as most flux/potential analogies). It has its limits, though.
>> Instead, I think it's more likely that the difference in DC voltage can vary significantly.
>
>Why do you think this?
Because as per the linked article, the absolute voltage potential of the earth in a local area can vary significantly over time. This make me think that the difference between a small isolated system and the earth ground must vary significantly. As you point out, everything actually is coupled, there is a time constant involved, but at least immediately after a nearby lightning strike you should have a significant difference. Because of the small capacitance of the isolated system, I'd expect you might see a high initial value that would soon (scale depending on the meter and the capacitance) decay to zero. Do you disagree?
> House wiring is not limited in how much current it can put through you
Absolutely agreed. Per the article, I'm referring to the supposed safety of an isolated transformer used a "razors only" outlet. In addition to having an intentionally small transformer, this system also has an intentionally very small fuse.
> The reason why touching one wire is “safe” is because the return path for the current is usually very high resistance
Yes. I guess the part that I'm failing to make clear is that I'm asking about the articles differentiation between isolated transformers and US standard circuits. In both cases, in normal dry clothed circumstances, there is normally a high resistance to any return circuit through your body. I'm still having trouble seeing the difference between the safety of the isolated system and the standard system. Most of the answers here (including yours) would seem to apply equally to both cases. I feel I understand this, just not the way in which the article explains the safety of isolated transformers.
I've gotten some helpful answers, but it's apparent that I've failed make clear my actual question, which is about the relative safety of a system with an isolated ground, such as the "razors only" plugs discussed in the article. The article seems to claim that such systems are safe for any single point of contact.
Maybe I can make it clearer with a parallel more specific question. Assume I'm running an gasoline powered generator which produces 120VAC for a single outlet. As far as I can tell, this is an ungrounded isolated system analogous to the isolated transformer. If I'm barefoot and standing on wet concrete, and stick my wet finger into only one side of the plug, will I receive a dangerous shock?
The article seems to claim that as long as I touch only one of the wires of an isolated system, I am "safe". My counter intuition is that that the capacitance of the earth is such that an AC current will flow through my body and cause me harm. Am I wrong? Is the article wrong? Or is there some essential difference between my proposal and the example in the article? If I'm wrong, I'd love to have a more technical explanation of why this seemingly unsafe behavior of sticking my finger in a live socket is actually safe.
> Assume I'm running an gasoline powered generator which produces 120VAC for a single outlet. As far as I can tell, this is an ungrounded isolated system analogous to the isolated transformer. If I'm barefoot and standing on wet concrete, and stick my wet finger into only one side of the plug, will I receive a dangerous shock?
Theoretically, no. Your generator isn't going to pick up large static voltage from the atmosphere the way a power grid does, so it's theoretically safe to leave it ungrounded.
However, I would strongly advise against actually doing this experiment unless you are absolutely certain that your generator is isolated from the ground, and even then I would advise against it. Some risks are not worth taking even in the name of science.
> However, I would strongly advise against actually doing this experiment unless you are absolutely certain that your generator is isolated from the ground, and even then I would advise against it.
A safe version of the experiment would be to take a standard incandescent lightbulb and run one wire to an earth ground and one wire to one side of the plug on the generator. My guess is still (although with more doubt based on how many intelligent people here have opined otherwise) that because of the enormous capacitance of the earth, the light bulb would still light up even though no clear circuit is formed.
In my quite possibly flawed mental model, electrons will be pulled and pushed from the earth across the filament, causing it to heat up. Whether this actually works (in my mental model) depends on the frequency and the capacitance of the generator, but I'm guessing that at 60Hz whatever is inside the generator is enough to get an effect. Unless someone has already done this, I guess I'll have to get a generator and try it.
> My guess is still ... that because of the enormous capacitance of the earth, the light bulb would still light up even though no clear circuit is formed.
Your statement implies that you are viewing the earth as a huge, perfect, capacitor. And by "perfect" I mean no series resistance.
In reality, the earth is a relatively poor electrical conductor (i.e., a high resistance [1]) and so where your mental model breaks is omitting that resistance from consideration in your model.
I.e., the equivalent circuit your 'experiment' produces would be:
G-----bulb------R--C
The "R" and "C" above are the capacitance and resistance of the earth. Yes, there is an amount of C present, but there is also a quite large series "R" present as well. And it is that resistance that results in a very limited current flow from your theoretical experiment. Too low of a flow to illuminate the bulb.
Your mental model is correct, but there's a matter of orders of magnitude, as I explained above. A standard incandescent lightbulb isn't going to light up until you have tens or hundreds of milliamps going through it, so it's possible it could still be dark at a current where you would be dead. At the microamp levels you'll actually see in this situation, you might be able to get an NE-2 neon bulb to light up. (Don't forget the ballast resistor.)
A good example are the helicopters with workers inside that work on high voltage lines while they are charged.
The workers touch the lines to bring themselves (and the whole helicopter) to the same voltage and then get to work. That high of voltage should be fatal, but voltage is relative. They are at the same voltage and there's no path for the electricity to follow.
In a truly ungrounded system, if you touch one of the wires in the system, your body will match the voltage of that wire and nothing "bad" will happen. Only if that voltage has a connection back to the other side of the system will current flow through you, either by touching the other wire or if one of the wires is grounded and you touch the "hot" side.
It's a great example, but a very different one. In the case of the helicopter, there truly is no path to ground. In the case of a person standing barefoot on wet ground, they (potentially) become the ground. If the human touching the wires is ungrounded (as in the case of the helicopter), there is no difference in this case whether or not the system itself is ground connected. The question is how much difference it makes for the system to be grounded if the human is also acting as a ground.
Note that I am not encouraging you to conduct this experiment :)
You are correct that there will be at least 60V that exists between your body and at least one of the wires from the generator. What isn’t guaranteed to exist is a low impedance loop for that current to flow through your body back to the generator. You’re in bare feet on the concrete, but the generator has a plastic body with rubber feet.
People repeat that old “it’s the current that kills you” line and often in a misguided and dangerous way, but it’s actually true in this case. If there’s no way for the current to flow back to the generator, the voltage on its own will likely not do much.
If you were to touch both wires while wearing rubber gloves, you’d be in a similar situation. 120V exists across your body, but because the gloves have a very high resistance, no current can flow (I=V/R)
"it’s the current that kills you" isn't the relevant point. The point is that the current through your body that kills you (specifically, the current through your heart, not merely through your fingertipes). Touching a high-current wire doesn't send high current through your body (same as high-voltage across your body) unless there is a circuit (as you correctly wrote).
Capacitive coupling within small portable generators is a bit of a rabbit hole. Send me an email and I can discuss.
Edit: kragen has good overview of this below from the generator in a puddle.
Now in practice, most small generators that you buy as a consumer are "floating neutral," which means that the generator chassis, EGC, and generator GSC (i.e. neutral) are NOT bonded together. Floating neutral generators are another, very long post.
---
Source: 6 years ago, when the NEC changed substantially for small portable generators, I wrote a multi-part series on all of this for a trade magazine in the pro audio industry, who make prolific use of portable power distribution systems, and generators, in the 3kW to 500kW size range.
In the course of writing that series, I dealt directly with the EE who works for PGMA, the portable generator manufacturer's trade organization.
So, it's true that the capacitive coupling of the system to ground allows some current to flow. If you connect yourself between one point in the circuit and ground in the way you describe, you become a resistor of about 1kΩ between that point and ground, and so you bring that point in the circuit to very nearly ground by virtue of flowing a current through your body. The question is, how big is that current?
It turns out that that current is precisely equal to the current flowing through the impedance between the other side of the circuit ("neutral", if you're connected to the "hot" side) and true ground. If we suppose that that impedance is purely capacitive, the question is just how big the capacitance is, because that tells us how big the impedance ((2πfC)⁻¹) is, and thus the current.
If it's 0.1 pF, the impedance at 60 Hz is 27 GΩ, and the current is 4.5 nA.
If it's 100 pF, 27 MΩ and 4.5 μA, which is still too small for you to feel. Megohms to tens of megohms is a typical resistance range for protection in the kind of neon-bulb testers that use your body to detect hot wires, because microamps are still enough to light up a neon bulb visibly.
If it's 0.1 μF, 27 kΩ and 4.5 mA. At this point you will probably yell and let go rapidly, but you will retain control over your muscles; this is still uncomfortable rather than hazardous.
If it's 1 μF, 2.7 kΩ and 45 mA. At this point your muscles will spasm and, depending on how you happen to be connected, you may be unable to let go until somebody else pulls you off. You will probably not die or suffer any permanent damage but it may be a few days before you feel okay again.
If it's over about 3 μF, and your body's resistance has become the limiting factor in the situation; for the purposes of electrical safety, it's not really any safer than if you had the generator grounded to start with.
So, the question is, how much capacitive coupling are you going to have between your gasoline generator and the actual earth, or rather, the wet concrete you're standing on? And this depends on geometry. Let's suppose the generator has a steel base, connected to its neutral wire, that is ¼m² and suspended on rubber feet of 10 mm above that same wet concrete floor. In this situation C = εA/d ≈ 220 pF, so you're down in the microamp range, three orders of magnitude away from an uncomfortable shock. If you instead have the generator on a dry wooden pallet or a dry concrete floor, you'll have about an order of magnitude less capacitive coupling. If it's on a dry wood floor elsewhere in the house and you're down in the basement standing on the wet concrete, you'll have about two orders of magnitude less capacitive coupling.
So in this situation what you need to worry about is not really the capacitive coupling; it's that maybe one of the rubber feet has a steel screw in the middle that has made conductive contact with the puddle.
Higher voltages make capacitive coupling more dangerous; you've probably seen the video of the lineman coming in on a helicopter to repair a live 100kV line, finishing by flying away while a meter-long arc connects his helicopter to the line. That's tens of milliamps of displacement current flowing across the helicopter's capacitance to the Earth, even though the helicopter is considerably further from the Earth than our hypothetical generator is from your wet concrete.
Yes, this is tremendously helpful for putting numbers on my intuitions. To understand the details, I'll need to think about it more tonight when I'm less distracted. As a starting question, does the capacitance internal to the generator matter, and what level might this typically be?
Also, how did you estimate the 1KΩ resistance to ground? Mostly as an aside, I presume you are familiar with the possibly apocryphal story of electrocution with a 9V battery? https://darwinawards.com/darwin/darwin1999-50.html
> does the capacitance internal to the generator matter, and what level might this typically be?
It might, but it can only make the situation safer. For example, if the "neutral" wire's only connection to the chassis is through its capacitive coupling to that chassis, then the impedance from that capacitance is in series with whatever the chassis-puddle capacitance is. Typically such capacitances are going to be much smaller, because the chassis and hypothetical puddle have much larger surface areas than anything else in the generator. Some can be larger, but their impedance isn't going to go below zero. (A sufficiently large tuned inductance could go below zero, but it would have to be very carefully tuned to lower the impedance rather than increase it.)
> Also, how did you estimate the 1KΩ resistance to ground? Mostly as an aside, I presume you are familiar with the possibly apocryphal story of electrocution with a 9V battery?
Mostly by sucking on ohmmeter leads in the past, although it's also common in textbooks. Yes, I've heard the story; if I had heard it earlier, I might have been more reluctant to suck on ohmmeter leads, although modern digital VOMs use smaller currents that are probably not dangerous.
> If there is 120V AC between the two wires, doesn't this mean that there has to be at least 60V AC potential between at least one of the wires and earth ground?
Yes, it does, but that's fine. The moment you touch a wire in an ungrounded system, that wire is connected to ground through your body, and the voltage between it and ground "floats" harmlessly and instantly to zero, with negligible current passing through you.
Practically, the system is always weakly grounded through mega-ohms of resistance in the insulation. Typically, this means that one wire is better grounded than the other (through a ground plane or part case), and it floats closer to one wire or the other. You can measure the voltage of the "ungrounded" system with a high-impedance multimeter or oscilloscope probe; I've probably done this more times by accident ("105 volts!?? Now 100 volts....Why is my motor ground at 95 volts and falling without turning the motor?") than on purpose.
Potential difference between two otherwise not-connected systems is not well defined. In practice the system will not be totally ungrounded, but there will be some high-impedance path to ground, and the impedance is hopefully large enough to not allow any significant fault current to flow.
I think this is getting close to the right answer, but I still don't understand the details. Let's say that we set up a large ground isolated system, and have it driving a resistance heater (20A at 120VAC). Standing barefoot on wet concrete, I tightly grab one of the leads to the heating element. With my other hand, I use a hammer smash the heating element so it's no longer a complete circuit. What happens and why?
I suspect that I'll get a terrible shock and likely die, because some substantial current will start flowing through my body. You seem to be saying that I'll be fine. Why do you believe this?
My guess is that the typical ground isolated system is safe not because of the isolation per se, but because the transformer used is small enough that the maximum current is not life threatening. When the circuit is intact, I'm unlikely to get anything other than a static shock. When the circuit is broken, it's current limited. My intuition is that this is the same as having a very small generator with only a few coils of wire. What am I missing?
> because the transformer used is small enough that the maximum current is not life threatening.
No, this is definitely not the case. 8mA is 1W at 120V, and although 8mA will not actually kill you, it will be a very nasty experience you will not want to repeat. 80mA, I wouldn't want to sell you life insurance. 160mA can definitely kill you.
Efficient transformers are also not effective current-limiting devices; their coils do have some resistance, and their magnetics have some leakage inductance which limits the current, but these only produce a small voltage loss at their rated output. Say you have a transformer designed for a 50-watt razor (400mA at 120V, several times more than you can survive). If its coils are dissipating 10 watts, it's going to get pretty hot there inside the wall while you're shaving, so this means that its coil resistance is already impractically high for this application. (Think about how hot a 10-watt compact fluorescent or LED bulb gets, even though it's exposed to airflow.) Yet that's only 62.5Ω, which means that if you short out its output, you're going to get 1900 mA through the short, almost five times its rated output and 19 times as much current as is needed to kill you. Using a more realistic coil resistance gives even larger peak currents.
(Also, using 20% of the total circuit power means that the razor sees 20% lower voltage, which may cause it to fail to work.)
The bottom line is, you can't depend on the transformer burning itself up or otherwise limiting current to protect you. The correct answer is given in my other response above.
Special transformers like neon-sign transformers and arc-welding transformers do have sufficiently high leakage inductance to not burn up if you short out their output, but such transformers are not used for general power application. Generally their maximum currents are still far too high to provide much protection for squishy human bodies.
> When the circuit is broken, it's current limited.
When the circuit is broken .. by definition, there isn't any current flowing!
You need a loop somehow from one of the terminals of the transformer through you to the other terminal in order to get a current flow.
(There might be some transient effect caused by the inductance of the system within one half-cycle, which may give you an arc as you smash the heater, but that can't continue and is unlikely to pass through you)
> If there is 120V AC between the two wires, doesn't this mean that there has to be at least 60V AC potential between at least one of the wires and earth ground?
One specific wire is “safe.” One wire (neutral) is at 0V, and the other wire (hot) is at 120V. So it is “safe” to touch the neutral wire. In practice, there are various fault conditions which will cause the neutral wire to become unsafe, which is why we now have a separate earth ground.
Perhaps an ignorant question, but given that when I travel I have to bring a seemingly endless array of plug adapters, which power standard is the "best" and how do we get rid of the rest?
> which power standard is the "best" and how do we get rid of the rest?
Not sure what's 'best', but getting rid of the 'rest' is likely to be expensive in terms of capital investment and for a comparatively minimal benefit. To put it from a US perspective, I have a house built in 1940, and while most of it's been rewired to more modern standards, there are still a number of places where it has the original 80 year old wiring. What this means is that I have devices I can't plug into all sockets in the house, and even this inconvenience hasn't been worth the thousands of dollars and personal disruption it would cost to upgrade those circuits. There are cheaper workarounds and more pressing uses for the time and money it would take to make the change.
It's this sort of reasoning that makes me hesitant to think that a unified global power standard is ever likely to happen, and that's even before considering some of of the infrastructure costs of making a change.
That said, modern flexible power supplies and dual voltage appliances seem like they go at least a large chunk of the way to make the situation at least a little better.
As far as I know, there's no advantages between 50 and 60 hz. I suspect the difference was political, but maybe someone else knows more about it than me?
As far as voltage is concerned, there's a very good reason why most of the world uses about 240 volts for "normal" outlets. The American 120 volt system is fine for things like TVs, clocks, computers, ect, but it's horrible for things like space heaters and electric cars. When I see videos from countries with higher voltages, I see much larger space heaters than we can handle on our outlets. Also, electric cars charge twice as fast from an ordinary European outlet than a normal American outlet.
The cost of copper even starts to make high voltages look better. In my house I have extremely thick (expensive) wires running to my heat pump, but if I had higher voltages these wires could be smaller (and cheaper.)
Transformers for 60 Hz are smaller and cheaper than transformers for 50 Hz. In general this continues to be true as you increase the frequency, although there are some effects that you need to compensate for as the frequency increases.
Your phone charger might have a transformer operating at something like 1 MHz. As a result, it can be made very, very small.
There's also a very nice property of 60 hz: Your clocks don't lose their time very often since they have a very predictable source of truth built right in.
I don't see a difference there between 60 Hz and 50 Hz. One has 1/60 of a second between cycles and the other has 1/50 so as long as you can count to 50 instead of 60 it's the same, right?
In Europe the grid phase is supposed to be pegged to UTC. In recent times (1~2y) we have some fluctuations because some south east european countries have some disputes about paying vs. being cut-off, which creates an energy imbalance in the network that's not trivially corrected for like normal small leakage or rounding errors. It is usually very stable over long time periods, considering the rarity of power outages lasting more than a few seconds. Most are just factions of a second, when the 10/30kV network (underground in cities) needs to be switched to a different upstream transformer. It's basically a circuit breaker with automatic failover. (Multi-second outages are explained by emergency load shedding, followed by re-connecting residential consumers and not the industrial ones. Necessary after long-distance transmission lines fail their redundancy and cause regions to be partially cut-off and then it's load-shedding vs. cascading failures.)
In the US the grid used to guarantee 5184000 per day with only short-term fluctuations (standard WEQ-006, the exact deviation allowed depends on which grid you are talking about). This was used by e.g. cheap clocks in coffee makers, but also by synchronous motors used by telescopes to align the telescope with the stars.
More recently, with our larger power grid, it is more efficient to deviate from 60 Hz. I do not understand the exact reasons, but for some reason the 5184000 cycle/day guarantee turns out to come with some cost when you have a hundred gigawatts of power spread across a thousand miles. There have been two petitions in North America to remove TEC and allow synchronous clocks to accumulate error, I believe this is going forward but not yet implemented.
> As far as voltage is concerned, there's a very good reason why most of the world uses about 240 volts for "normal" outlets.
If North America would ever decide to start using 240V for appliances (through a transition period), what style of plug should be used? UK-style (Type G)? Europlug? Other?
I'm a great proponent of the UK (Type G) plug. It's wonderfully over-engineered for safety.
The plug is an asymmetric triangle shape, with a well-defined orientation that is "forever" polarised. The earth pin (at the top) makes contact first, as it is longer and physically larger. In the process of doing so, it opens two shutters on live ("hot") and neutral within the socket. The live and neutral pins themselves are insulated from the base of the plug to a distance slightly more than the extent by which the earth pin sticks out. Therefore, it's impossible to make contact with a live exposed pin, even if you drop a pin in the gap between the plug and the socket. Each plug is individually fused, with an amount that varies by each device. Cable strain relief is integral, and the wire lengths are designed such that if someone pulled the cable out incredibly hard, live breaks first, followed by neutral, followed by earth. The plastic is designed to char and not melt if burnt, and a standard 13 A (240 V) plug can take around 50 A at 240 VAC or instantaneous kV before coming a-cropper. The contact forces on the pins are far higher than a US plug. This means that a plug is very unlikely to arc in its socket, leading to less wear and tear and longer lasting plugs and sockets. I like the fact that each socket is individually switched, which further ameliorates the above. Because of the higher voltage, the maximum commonly used residential current, 13 A, delivers ~3 kW of power as opposed to the US maximum of ~1.6 kW, so you don't need a further set of incompatible plugs.
Still hurts like hell if you tread on one barefoot though.
Most houses in the US already receive 240 volts, though a biphase system, so that you can tap just one of the phases and get 120 volts.
But you can easily connect to both phases and get 240 volts, and this is quite common for appliances such as ovens, clothes dryers, and now EV chargers.
Assuming you read the article, the difference is that they have two hot prongs instead of a single hot pong. The 4-prong outlets, like a standard dryer outlet, allow a device to use both 120v and 240v. Sometimes appliances take advantage of this by using a 240v heating coil and a 120v motor.
Adopting EU-style plugs is more problematic. There's the two hot prong problem, instead of one hot and one neutral. The other problem is frequency, as we use 60hz instead of 50hz.
EU-plugs don't treat hot and neutral side differently at all, it's just a convention and in practice it doesn't matter when they are reversed, it does happen often. Only the earth must be always correctly connected. It would work fine with two hot prongs too.
I think the Australian plug is great, with the only caveat that it would be basically perfect if the pins were just slightly thicker (they're just a little bit too easy to bend).
It's fairly compact, sockets always have ground, you can't plug it in upside down and swap live/neuteral with the two pin version (used for double insulated equipment), and it has a sensible scheme for 10, 15 and 20 amp sockets (the ground pin is bigger for 15 amps and all the pins bigger for 20 (I think) so you can plug a 10 amp plug into anything but can't put a 15 amp plug in a 10 amp socket or a 20 amp in a 15 amp socket).
In my experience, the Italian/Chilean Type L has by far the most convenient geometry of these plugs, with a very compact design as well as modern safety features. Main downside is that it is not polarized.
Schuko is pretty solid choice, and conveniently already being used by large part of Europe. If polarization is must for some reason and schuko feels too bulky then I guess IEC 60906-1 is as good choice as any.
I've always wanted an international standard with notches that indicate the voltage and frequency. A device that accepts all voltages and frequencies will accept any notch, a more selective device would block a notch for the frequency / voltage it can't handle.
(And the plug would need to fully cover the pins once they have current. That's the big problem with the NEMA and Europlugs, the pins are still exposed when they're live.)
I'd like to see IEC 60320 C13/C14 which is already common with computers, it is small, easy to use and properly covers the pins. But perhaps it isn't as mechanically robust when connected and disconnected often?
I hate Schuko plugs. They're circular, but there's only two correct orientations you can plug them in, which results in a lot of finagling if you don't have a good visual lock on the outlet. I wish they were rectagular.
Having non-polarized sockets forces devices to be polarization agnostic, which could be safer... Not that I actually know anything about electric safety engineering. And of course there is the ergonomic aspect that allows you to insert the plug in any orientation, compare to usb-c
I may be biased, but I think British plugs are great. They have integrated fuses and are switchable too! The sockets have shutters as well to stop children sticking paperclips or forks in and electrocuting themselves.
The British system (BS1363) is by many measures the best plug/socket system, from automatically shuttering the live wires, to insulated live/neutral to prevent metal objects from shorting on sockets, fuses in the plugs to protect weaker cables, longer wires for Earth than Live if the flex is pulled out. Even 1363 has flaws though -- see the loophole at http://bs1363.fatallyflawed.org.uk/, and of course the counterfits.
Every time I use a U.S. plug/socket is scares the bejesus out of me, having grown up with 1363.
However it has significant practical drawbacks. Many devices are class 2 and thus down need earthing. A flat two-pronged europlug allows a far higher desnsity socket, and much smaller footprint cables.
BS 1363 plug has some neat safety features, but it's really big. Look at the size of a British power strip, for example. As Tom Scott observes, to get the full benefit of the safety features, they would need to make them even bigger. It's also not even close to round, so pulling it out of a nest of cables looks terrible.
I'm not convinced that a fuse belongs in a plug, either, especially if it's invisible, and the plug has to be completely disassembled with a screwdriver to access it. Put the fuse in the device, which certainly has much more space for it, and probably a mechanism to make it easy to access.
As far as the physical structure goes, BS 1363 looks like it was designed in the 1940's ... because it was designed in the 1940's. I'd much rather have something modern and ergonomic like TRUE1. It's small and round, it locks, it's basically impossible to shock yourself with it, and it's IP rated.
I'm not alone here. There's many videos on YouTube of people hacking the Edison plugs off their power tools to replace them with powerCON or TRUE1, but I have yet to find anyone in the world who wants to replace their Edison plug with BS 1363!
>I'm not convinced that a fuse belongs in a plug, either, especially if it's invisible, and the plug has to be completely disassembled with a screwdriver to access it.
Moulded plugs generally use a clip-in fuseholder that can be opened with any pointed object; rewireable plugs can be opened to replace the fuse by loosening one captive combination Philips/flat screw. It's usually far easier than replacing a fuse in the appliance, which is often on a PCB-mounted holder buried within the guts of the appliance.
The fuse is located in the plug, rather than the device, because one of the most likely ways to be exposed to live wire is a damaged/frayed power lead. If the fuse is located in the device, it does not protect the upstream part of the system.
I just got through adding outlets and lights in the garage this weekend, but I’m otherwise kind of a dumbass about such things. Should I have to choose, though, I’d go 240V at 60hz. 240V so my coffee water heats faster than the 120V we get in the US, 60hz just because you gotta choose a number, and 60hz makes for good wall clocks.
After a recent power outage, we found that a seemingly random selection of sockets and lights had become prone to intermittent glitches. By popping one circuit breaker at a time, I realized that these circuits coincided with breakers on the right side of the box. From this, I found out what is apparently a common pattern: we have two 120V out-of-phase incoming lines together with a grounded neutral that is presumably the center tap on the secondary winding of the local transformer. Between them the two live lines give 240V, which is used by the AC unit.
Unfortunately, while you may have the voltage for a faster coffee maker, you may not have the wiring in your kitchen and I don't suppose you can find a UL-approved coffee-maker for that set-up, which has two live wires (though maybe they are made for restaurant and institutional kitchens?)
> I realized that these circuits coincided with breakers on the right side of the box. From this, I found out what is apparently a common pattern: we have two 120V out-of-phase incoming lines together with a grounded neutral that is presumably the center tap on the secondary winding of the local transformer.
That's exactly right... this sort of split phase approach is common, at least in the US. Larger consumers can have 240V available to them and everything else is 120V. (Larger consumers can mean things like ovens, ranges, wall AC's, normal AC's, larger shop tools... the list goes on and on.)
Split-phase also has a couple interesting failure modes. A few years ago, we had a case where a 120V heater would cause fans in other parts of the house to speed up when it switched on. What was happening was that the neutral wasn't fully grounded, so when the heater switched on one phase, it pulled the neutral's voltage in a way that caused the other phase's voltage to go up. Once the electrician figured that one out, it was fixed with the quickness. (Not safe or good for electronics at all.)
If you want to get a visit from the power company as urgently as possible, just call them and say the phrase "floating neutral" (of course, they'll be mad at you if it's not true).
Our neighbor's (poorly-attached) service mast came down a few weeks ago, causing a short that burned out the shared wires that connect both them and us to the transformer. After the initial event (complete with loud pops, sparks, flames...), our power actually came back on; but I insisted on keeping our main breakers turned off until the power company made their repairs. Turned out to be a wise move, as we indeed had a floating neutral.
What's doubly fun are multiwire branch circuits. Basically, instead of splitting the 240 V into 120 V at the breaker box, you take a 240 V branch circuit and split it into two somewhere down the line. This has two advantages: first, you need fewer wires going all the way to the breaker box; second, resistive loads on opposite legs of the circuit cancel each other out on the neutral, so you can (in theory) run 30 A worth of stuff off a single run of 14-gauge 4-wire Romex (= cheaper).
Anyways, where the fun comes in, is if for some reason the neutral gets disconnected before the branch split. You now are effectively running the appliances on one branch in series with those on the other. Depending on the particular appliances involved, this means you could see up to 240 V on one leg of the circuit. This is, of course, very bad.
I lived in a place where the ceiling light would only turn on if one of the stove plates were on. If I recall correctly, the stove plate would not get hot in that setting.
In a theater I worked at, I discovered that one outlet was accidentally built using the two live wires instead of neutral and hot. Unfortunately, I discovered it while testing a lighting instrument, promptly blowing out its $30 lamp. (And like a proper scientist, checked to see if it did that every time, blowing out another $30 lamp.)
Once I stopped destroying expensive equipment, it did amazing things for my circular saw. (Probably not good for that, either, and fortunately we promptly moved out.)
I have the necessary wiring in my kitchen already. The volts went up, not the amps. I believe you’re assuming that I’m running a US dryer plug or something. No, the question was “which standard?” The assumption being that in this hypothetical world my appliances get upgraded, too. IOW, I wake up and I’ve got European 220V at 60hz with US plugs.
>By popping one circuit breaker at a time, I realized that these circuits coincided with breakers on the right side of the box.
I hope you got this checked out! There could be something wrong with the phase going to those breakers (if your box is of the kind that has phases split by side which I am unfamiliar with).
Volts make no difference to heating your coffee. That is all about watts which is mathematically volts times amps. Lower voltage, higher amps for the same time to heat your coffee.
The higher amps trips the circuit breaker. If you don’t think a difference in voltage determines one’s tea water heating time, check the wattage rating on a kettle sold in the US (120V) and one from the UK (240V).
W can be constant, in which case there is no contradiction. Of course then V=IR means I need to change R as well. However this is all math. You can choose to hold any one factor constant and change the others to fix your needs.
In the real world V is a constant outside your control (120 in the US, 220 elsewhere), and A is limited by local wiring conventions (which means realistically the amount of watts you can get is different - but this is not a factor of voltage). The only factor you can change is R, which turns out to be easy enough (in this application).
I like the Swiss ones, because they get the three prongs in a more compact design than the bigger European round plugs, and are easier to pull from the socket than the harder force needed to pull European ones. They Swiss 3-prong plug is as large as the European 2-prong one.
In a wall outlet the size of a European single socket (outlet?), the Swiss one gets three sockets.
The UK one is supposedly safer, but it's enormous...
It's simple, you just refuse to produce anything with an alternative standard, perhaps even banning the export of other standards citing safety concerns, then as more countries find themselves using largely your products with adapters they will eventually realise it's easier to adopt it and use adapters for their native legacy standard.
I like IEC 60309 best. 480V 30A three phase IP68/69k rated plugs and sockets? No problem with pin-and-sleeve connectors. Any food processing plant you go to will be full of them. Probably not a great choice for a table lamp, though.
I'm a big fan of Schuko plugs/sockets. Not gigantic like the UK plug, and the sockets grip the plugs really well, plus there are two ground contacts for redundancy.
Higher voltage allows for lower amperage for given watts / more wattage at a given amper limit. The higher the amperage, the larger the wires need to be to avoid excess heat. In the US, standard household circuits are rated for 15 or 20 amps, in the EU 10 and 16 are common. So, you can expect a 3.5kW AC unit to work without problems in the EU while getting anything more than 2kW in North America is a serious headache.
Where you can put either three Europlugs or 1 Europlug + 2 of either S11 or S17 OR a single Schuko (and it fits in 2 modules, like a "normal" schuko or schuko multistandard)
The article isn't not really on point. A two-prong system still has an earthed neutral. The third prong is purely for redundancy. If neutral happens to be disconnected, there is still a return path. It also protects against mixed up neutral and hot. If the chassis of a device is grounded to the third conductor, and that one is properly routed, then an inadvertent reverse wiring of hot/neutral is mitigated. A third conductor also makes possible ground fault detection: situations when a device is not returning all current through the neutral, because its safety ground has become energized.
How might a neutral end up disconnected? Not just by accident; for instance, someone would incorrectly wire a cold-side switch instead of hot-side (E.g. wall panel switch that controls an outlet) so that when the outlet is off, the neutral has been cut off, but hot is still live. In that situation, the third conductor would still be grounded. Nobody would mistakenly implement the switch on the ground line unless they didn't bother testing it, because the switch then wouldn't work; the ground line is implicitly protected from having switches accidentally planted on it by the fact that interrupting it does nothing, functionally.
Three prongs was a good idea before the existence of ground fault interrupters, but now that they do exist, ground fault interrupters with two prongs provides a much better solution.
The problem with ground is that it's a conductor. If there's a fault on it, you don't know exactly what it's connected to, so now you have a situation where the metal chassis is connected to something. You touch the chassis and the really-grounded a sink and you have a problem.
Here's a real world example of this. The ground is bonded to neutral at the panel. But what happens when neutral between the pole and the house breaks? This happened to my neighbor. There is still 240V with neutral/ground floating somewhere in the middle. All of his appliances on the less loaded side of the 240V blew up and "ground" was very not at earth ground.
That's a problem with incorrect installation / code. In a proper TN-C-S system the combined PEN will be split up at the main panel, and PE will be local ground and pipework is typically connected to that.
If GFI sockets and breakers had been widely available at the time, would the third prong have been added? Would a GFI along with the "one long slot" plugs make the ground wire obsolete?
GFI outlets are $15+, and normal outlets are $2. This adds up when you're building a house: a typical house has something on the order of 30-50 outlets, so you're looking at $400-$700 vs $60-$100. It might not seem like much in the grand scheme of things, especially to software engineers making lots of money, but if we had this approach not only for outlets, but also for everything else, building literally anything would cost 5 times as much (and in fact it does these days, to a large degree because we can afford spending 5 times as much on building).
Where are you getting your prices? Last time I checked, GFCI outlets were indeed around $15, but the normal ones are less than a dollar when you buy them in bulk packs (which you would for building a house). Otherwise, your point is sound, but the price difference is even greater than you said.
Those standard outlets are really dirt-cheap, and it's annoying that builders are so cheap about installing them. New construction has gotten better though, but really there should be an outlet every 5-6 feet in my opinion, or maybe even less.
Home Depot non-bulk prices -- the only ones I'm familiar, as I haven't had a need to buy many of them at once.
Also worth noting is that you can wire your circuit in a way that the first one is GFCI outlet, and the subsequent ones are daisy-chained to it, so that all of them are protected by GFCI. Of course, there are reasons to not do it (e.g. it is then quite confusing when the GFCI trips while using some non-GFCI outlet, because it might not even occur to people to check the GFCI outlet which might not even be used), but that's also an option if you want to save money on outlets.
Also, I think the reason for few outlets in buildings is not the cost of the device, but rather labor required to put it in. Connecting wires to the outlet is quick, but you need to also install boxes, route the wire to the boxes, mark and cut the drywall to get to these boxes etc. It can all add up to significant labor time, and if you look at what electrical contractors charge, the cost of the device itself is a small part of it.
>Home Depot non-bulk prices
Definitely check out the 10-packs; if you own a house that isn't new, and are already doing some wiring work, it's not a bad idea to keep some extras around and maybe replace some of the crappy older outlets when they're so cheap.
>Also worth noting is that you can wire your circuit in a way that the first one is GFCI outlet, and the subsequent ones are daisy-chained to it,
Yes, this is actually normal for kitchens, where all the counter outlets are on the same circuit. Just put the GFCI on the outlet nearest the breaker box, and all the others are then protected by it. It is really annoying, however, when people retrofit these things into old houses and you get something like a downstairs bathroom outlet that has no apparent GFCI, but then it trips, and you have to hunt around the house for the GFCI outlet only to find it in a bedroom on the upper floor. I rented a house for a while where both bathrooms and both bedrooms were all on the same circuit (lights too!), and on an old GFCI outlet that would frequently trip and leave me in the dark. Extremely unsafe.
>I think the reason for few outlets in buildings is not the cost of the device, but rather labor required to put it in.
Yep, that's exactly the reason, but it's still annoying because years later, you have to move furniture around to find an outlet that your device's cord can reach because there just weren't enough installed.
> It is really annoying, however, when people retrofit these things into old houses and you get something like a downstairs bathroom outlet that has no apparent GFCI, but then it trips, and you have to hunt around the house for the GFCI outlet only to find it in a bedroom on the upper floor.
They do have GFCI breakers. I'm not sure how they compare in cost, but one disadvantage is that you can't easily test them (who ever bothers going to their breaker box to test the GFCI occasionally?), nor can you easily reset them when they do trip.
Probably. AFAIK GFI measures the current difference live and neutral, so a current between live and ground (as in the third slot) would trip it. Imagine if someone forms a path for current between a "grounded" metal case and live. If there is no third slot then "grounding" means connecting to neutral, so GFI wouldn't trip.
Possibly, GFI mechanisms are good, but they're not as good as a true ground since they take time to activate and require a minimum current draw to trigger (you can still be injured or killed below that current draw)
What I learned from reading every single one of the 184 comments (as of 6:28 pm ET) on this post: call an electrician for anything more complex than changing a light bulb.
For some reason, this reminds me of the Laryngeal nerve. It's like a natural process repeating itself in different systems. Because re-engineering the solution from the ground up to be more efficient isn't an option, there has to be small incremental steps to arrive at a solution that works.
> Because of the inefficiencies of the routing the nerve takes, it's often hailed as one of the most striking cases against intelligent design — or at least calls into question the "intelligence".
Man, what an obnoxious way to start an article.
EDIT: link was removed, this comment no longer needed.
Oh my bad, didn't mean to suggest it was intelligently designed. I didn't read that full page and didn't pick up on that context. I'll remove that link.
"The sparks occur because of a little-known fact: all the world is a gigantic electrostatic generator. There is a flow of charge going on vertically everwhere on earth. Thunderstorms pump negative charge downwards, and the charge filters upwards everywhere else on earth. Depending on the height of your circuitry above the earth's surface, depending on the area covered by your wires, and depending on whether there was a thunderstorm above you at the time, there might be a fairly huge DC charge on your electrical distribution system. This charge might be several hundred volts; enough to zap computers and delicate electronics. Or... it might be many tens of thousands of volts, enough to create enormous sparks which jump across switches and leap out of wall outlets, wall switches, across transformer windings, etc. Your electric power system is acting like a sort of capacitive "antenna" which intercepts the feeble current coming from the sky and builds up a huge potential difference with respect to the earth."
For those outside the US, or generally confused how US domestic power distribution works, let me explain in the context of the 3 prong plug and NFPA 70 NEC (National Electrical Code) for the lay reader. We'll start outside the house and go all the way to the plug:
0. Before the transformer secondary on the pole outside the house, the wiring is dictated by the NESC, which is like the NEC for utilities. We'll leave that there.
1. Most US homes have a single phase, "center tapped" transformer secondary. This gives three voltage potentials to feed the home: (Nominal) 0V - from the center tap; +120Vrms - from one "end" of the transformer; -120Vrms from the other end of the transformer. And, yes, I am glossing over the phase relationship here. All three potentials are connected through the chunky copper of the transformer, which provides a current path to complete the circuit.
2. The three potentials feed the house meter and disconnect. The 0V potential may, or may not, be insulated on the path back to the transformer secondary, and often doubles as the mechanical support for the incoming lines (the "service"). The other two potentials will be insulated wiring.
3. Inside the main service panel the +120V and -120V inputs are split between the breakers. Every other breaker is connected to +120V or -120V. Typically a breaker that spans multiple slots is connected from -120V to +120V. That is the "240Vrms" for ovens, hot water heaters, dryers, air conditioners, etc.
4. Adjacent to the metal bits that fan out the +120V and -120V is a strip of metal that allows multiple connections to 0V potential. Colloquially this the "neutral bus," but in the NEC this is termed a "grounded service conductor." All of the white wires from the circuits in the house ("branch circuits") are tied to the spaces in this metal bus bar. This metal bar is also electrically connected to the 0V potential entering the house from the transformer center tap.
5. Also inside the main service panel there is strip of metal that allows connection for a number of "grounds." It looks just like the "neutral" connection bar, but instead has a number of green wires attached to it. Colloquially these wires a called "ground," but the NEC calls them "Equiment Grounding Conductor" or EGC.
6. Between the grounded conductor ("neutral") bar, and the EGC ("ground") bar there is a removable conductive link. In the main panel this link remains installed, but in secondary panels it is removed. The link is removed in "subpanels" to insure correct operation in the event of a fault. Fault conditions are discussed below.
7. Also connected to the EGC and grounded conductor is a third wire. The NEC terms this the "grounding electrode conductor" or GEC. The GEC then goes to a water pipe or "grounding" rod(s). Thus the GEC is the connection between physical Earth voltage potential and the power panel. It is unfortunate that the EGC and GEC are such similar abbreviations.
In the event I'm out of characters, I'll pause and reply to myself now.
Brief recap: Three different voltage potentials in from the pole on three different conductors. And in the main panel a link between three other wires: the EGC ("ground"), GEC("earth"), and grounded service conductor ("neutral").
The next section is where people go awry, even licensed electricians I have met. But it is also the meat of the safety aspect of the three wire configuration. We will assume a house with a single electrical panel for simplicity.
8. Three wires traverse from the main panel, down a branch circuit to the three prong "edison" outlet on the wall: One of the three prongs connects to either +120V or -120V; another connects to the EGC bar inside the panel; the last connects to the grounded service conductor bar inside the panel.
9. The short blade on the top row of the Edison connector is +120 or -120V; the tall blade on the top row connects to the grounded service conductor (colloquial neutral). The single connector on the bottom row connects to the EGC (colloquial ground).
10. If you plug in an item that has a three wire plug, the lower single plug prong will be longer. This is so that the item is electrically connected (i.e. "bonded") to the EGC before the other two wires. This insures the "fault current" path is connected before the item is energized.
11. Under normal conditions, the current path is through the item plugged in between either +120 or -120V and 0V volt potentials. Crudely think of current coming "out" the short plug prong and "in" the tall plug prong. The EGC (colloquial ground green wire) doesn't do anything under normal operation.
12. The current path is then back down the branch circuit, via the grounded service conductor (colloquial neutral). the metal bar that links all the grounded service conductors together then has a path back to 0V on the transformer by one of the three incoming conductors.
13. Finally, the current can flow from the 0V location on the transformer to the higher potential via the wire of the secondary. Notice that the GEC (earth) connection to the ground rod was NOT a meaningful component of the current path.
14. Returning to #11, and considering abnormal operation. Here current somehow flows outside of the correct circuitry in the powered item. The ECG is bonded (connected) to the item's chassis and provides an alternative current path. The EGC provides this connection back to the panel bar with all the green ECGs tied to it.
15. The EGC bus bar in the panel has a conductive link back to the grounded service conductor bus bar (colloquial neutral) via the removable link that we discussed previously. The link then "brings" the current over to the grounded service conductor (0V potential from the street), and provides the current path back through the transformer secondary.
16. Let's assume the outlet is a GFCI / RCD. It notices the current coming back on the grounded service conductor doesn't match the current going "out" into the device, and opens the circuit. That is because the fault current is "lost" to the EGC, and goes around the GFCI outlet. The outlet trips, assuming a human body is the fault current path.
17. People commonly assume that the GEC (earth) conductor somehow matters for current path in event of a fault, but this is rarely the case. Usually the water pipe or grounding rod(s) are high impedance relative to the wire in the transformer secondary, and so the current divider formed is essentially all through the grounded service conductor.
18. The GEC (earth) connector holds the pole transformer secondary center tap near the same relative 0V potential as the house. The GEC may also provide the lower impedance path at very high frequencies encountered during a lightning strike.
If you made it this far, I'm happy to field further questions :-)
"If a GFCI only cares about the current imbalance, and the traditional style breaker opens to due to a big spike in current, why care about what path it takes through the building wiring back to the pole transformer?"
If a sub panel does not have the jumper removed, then the EGC and GSC between that sub panel and the main panel will form a current divider. Let's assume they are the same wire gauge, basically the same resistance, and so essentially split the return current.
This leads to a couple problems:
1. The EGC may no longer able to carry its full rated current load, due to the baseline "normal" current that the EGC is also bearing. You want to bear as much current as possible to quickly trip a conventional breaker in event of a fault to the EGC. This is real, but relatively minor concern.
2. All EGCs "upstream" of the bonding point have been "contaminated" by the normal neutral current that got on to the EGC at the mistaken sub panel bond. This could "put current" on the surfaces of other equipment, exposed ground wires, conductive raceway, etc. This is the more major concern, as these surfaces would have the full voltage potential available to them should there be yet an additional current path.
In the main panel the EGC, GSC, and GEC are all connected at a single point. And from this point there is only one path back to the transformer, via the 0V conductor of the GSC. The transformer side of this 0V potential conductor will match the local Earth voltage, set by the GEC, and wiggle from 0V nominal by the fluctuations caused by current (V=IR) in the GSC conductor between the panel and the pole.
BTW, if you have a portable generator for emergency purposes and it's connected via a transfer switch, take note of the important grounding and bonding requirements:
These are called CLASS II devices and they are designed with double or reinforced insulation. In plain terms it means that the current carrying components are themselves insulated AND surrounded by a second layer of insulation. That or the current carrying components are protected by reinforced insulation like a hard plastic case.
A practical example is the humble power brick or wall wart. These are class II power supplies which are enclosed in a hard plastic case which protects the user from electrical shock.
The only 2-prong device I own is a (Japanese-made, of course) rice cooker. You'd think that if any device warranted 3-prong safety, it'd be a metal container which combines water and mains electricity.