The Electrical Outlet and How It Got That Way

Right now, if you happen to be in Noth America, chances are pretty good that there’s at least one little face staring at you. Look around and you’ll spy it, probably about 15 inches up from the floor on a nearby wall. It’s the ubiquitous wall outlet, with three holes arranged in a way that can’t help but stimulate the facial recognition firmware of our mammalian brain.

No matter where you go you’ll find those outlets and similar ones, all engineered for specific tasks. But why do they look the way they do? And what’s going on electrically and mechanically behind that familiar plastic face? It’s a topic we’ve touched on before with Jenny List’s take on international mains standards. Now it’s time to take a look inside the common North American wall socket, and how it got that way.

Hubbell’s Plugs

Separable Attachment Plug, US Patent 774,250. Note the round, headphone-like prongs rather than flat blades.

Consider the problems faced by engineers and designers in the early days of the electrical age. They were literally inventing an industry from the ground up, with very little to go on in terms of prior art. Not only did they have to invent the means of producing electricity, they had to come up with absolutely every component that would connect together to create useful circuits for paying customers, preferably without killing them.

One thing customers, particularly residential customers, would need would be a means to temporarily attach electrical devices to the mains supply, without requiring a visit from an electrician to connect them to the fixed wiring of a house or office, which was typically dedicated to sockets for light bulbs. The requirements were simple: provide two contacts, one for the line conductor and one for the neutral, that could remain firmly connected but easily interrupted at need.

Imaginative minds worked on this and similar problems in the late 19th and early 20th centuries, and various solutions were adopted. But it wasn’t until 1903 that Harvey Hubbell, an inventor from Bridgeport, Connecticut, patented his “Separable Attachment Plug,” a device that we’d recognize as a plug and socket. Hubbell’s first pass at a design used round conductors that looked a bit like the plugs used in manual telephone exchanges to make connections, and might have been inspired by them. The detents at the tip of the pins were retained by the spring action of the contacts inside the socket.

A Hubbell plug with flat blades, from the 1905 catalog.

The device worked well, but the manufacturer and businessman in Harvey saw problems. Foremost was the costs behind those round pins, which would have required machining to achieve the tip and detent. Harvey would have known that parts stamped from sheet metal would be cheaper and easier to manufacture, and so he scrapped the round pins in favor of flat metal blades in 1904. Like the round prongs, the flat blades had a detent for retention, and were arranged in a line. Catalogs from the time list dozens of variants of the “Hubbell Attachment Plug,” and the prices shown for each device suggest that Hubbell’s company fared well in the early 20th century.

For reasons unknown, though, Hubbell altered his design in 1912. The two blades were no longer in a line; each blade was twisted 90° to form the familiar parallel arrangement we see to this day. Hubbell continued to sell both styles of plugs and sockets, and by 1915 had sold something like 15 million units, enough to ensure that Hubbell’s design would be adopted as a standard, even without the millions of units also sold by Hubbell’s imitators.

Standardization

The specifications for the standard wall outlet we know and love today in North America are determined by the National Electrical Manufacturers Association (NEMA). NEMA standards cover a bewildering range of electrical products; we’ve covered their enclosure and weather-resistance standards before. The standard 120-volt, 15-amp outlet is a NEMA 5-15. The third conductor, the ground pin that completes the outlet’s face, is a round or U-shaped prong. It was added to some outlets as early as the 1920s as a safety feature and is now required for all outlets by the National Electrical Code.

The ground connection is interesting. You’ll notice that on three-wire plugs, the ground pin extends further out from the insulated cord body by about 1/8″. The idea here is that the ground circuit will be completed before the line and neutral connections are made when plugging the cord into an outlet, and perhaps more importantly, will be disconnected last when unplugging. That ensures that there’s a path to ground any time a circuit is plugged into the outlet.

Note too that the NEMA standard says the ground pin is actually located above the slots for the line and neutral pins, turning that frowning face upside down. There’s some logic to that — if something conductive should drape across a partially unplugged cord, it’s safer to have the line and neutral blades physically blocked by the ground pin. In practice, though, most outlets in residential and business settings are installed with the ground plug down. But look around the next time you’re in a hospital; chances are, the outlets there are all installed the correct way.

Behind the Face

The internals of a NEMA 5-15 outlet vary by manufacturer, of course, and even within a brand, there are different grades of outlet. The picture below shows two different grades of outlet taken apart. They’re similar in that both the line and the neutral connections are formed brass bus bars, with screw connections on the outside for connection into a building’s wiring, and springy contacts to grip and retain the mating plug. The industrial-grade outlet has thicker bus bars, better contacts, and stouter plastic in the body. You’ll notice too that both grades have the ground pin directly connected to the metal frame of the outlet, which would also be in contact with a metal wall box, if it were mounted in one.

NEMA 5-15 outlet internals. Source: HandymanHowTo.com

Considering how much else has changed in the last century, it’s pretty remarkable that Harvey Hubbell’s original plug and socket designs have remained pretty much unchanged. They’ve been tweaked, for sure, and the original idea has been extended to a panoply of configurations for every connection imaginable. There’s no doubt that the design has some deficiencies, but in the end, Harvey’s ideas seem to have won the day by addressing the basic needs.

195 thoughts on “The Electrical Outlet and How It Got That Way

          1. The British plug is definitely not the best. First of all they should admit that their way of working (fuse in the plug) is outdated due to the copper shortage during the war. Now saying that that is the safest way of working is bullshit. You never know what kind of shitty fuse a Chinese cable has, so my main fuse of my house is at least safer than this.

            Secondly a plug that is tamper proof but can be bypassed by plugging in the plug backwards (connect only the earth pole) is by far a good design. EU, or US plugs that have the tamper proof system based on the Live and Neutral prongs is much safer.

            The only criticism on the US plugs is the lack of electrical shielding of the screws on the outside of the outlet and the lack of shielding of the prongs when they are inserted and making contact.

          2. @Bart
            I agree, you can’t rely on what fuses a Chinese plug has got, and the move to molded plugs with a non-replaceable fuse (which I think was first due to US companies who didn’t understand U.K. plugs) means you can’t check and replace the fuse. When I was a kid, we regularly checked the ampage of equipment and ensured the correct fuse was in the plug when switching plugs.
            There’s an argument that the plug fuse should be matched to the wire spec, not the equipment, and I believe Chinese manufacturers follow this fairly well.
            However, most equipment has its own fuse, and the circuit has its own breakers, so it remains very safe.

            But you can’t defeat the safety mechanism by putting he plug in backwards (I assume you mean upside down?) – the plug pin cannot engage the safety mechanism if you do this as it doesn’t protrude far enough.

            The main issue with U.K. plugs is their size and tendency to lie pins up on the floor. Electrically, they’re much safer. The residential US socket shown above looks like a toy compared to a U.K. socket.

          3. @BART

            Err we also have fuses, (acutally MCB, RCB, RCBO etc) as well. Also what do you mean plug in backwards? I can’t think of a way the UK plug can be plugged in backwards.

    1. I agree. Remarkable that it hasn’t changed much is the wrong word here. It’s similar to why many people still use QWERTY keyboards. Because everybody else does it and it is very difficult to change. Sort of like daylight savings time as well, though this is much harder to change.

      It’s still not a great design choice but it’s a codified standard at least for the US. The ubiquity of commercial and residential hardware with that plug basically requires that you use it as well.

      Not sure how you would even practically change that without a truly massive rewiring campaign and adapters and all sorts of incompatibility issues for many years.

      Are there other examples of legacy products that are actually somewhat harmful that have persisted as long as this?

      1. Except that “everybody” are actually using safe plug designs and a metric system. It’s just USA that likes to be shocked regularly and read all measurements in stupid fractions.

        1. Not true, Chinese plug standards also expose the conductors when being plugged in. Australian electrical code only mandated the insulating sheaths over conductors in 2005. Etc.

          1. They’re only easier to visualize because you’re used to it. If you’re used to decimal places, that’s much better. I recommend you start practicing and you’ll see what kind of advantages it has.

            What’s 1/16 meter plus an eighth of a centimeter? Quick. What’s 6.25mm + 1.25mm? Same thing. It’s pretty apparent which one of these is handier. By the way, that illustrates how power of two fractions don’t nicely line up with power of ten measurements past a quarter. And if you’re measuring things, you’ll find it’s much easier to just note the number of millimeters. Your meter stick probably doesn’t have fractional markings, and even if it did it’s much more likely that your measurement will be closer to a millimeter mark than a power of two fraction that isn’t absurdly small like 1/128. Why is it easier to visualize a measurement that isn’t marked on your tools, anyway?

            I guess another way to think of it is that one is a fractional system using power of two, and the other uses power of ten. Our counting system also uses power of ten, so unless you’re counting in binary it’s going to be easier. There’s almost nothing salvageable about imperial units except the fact that a few people are very used to them. Take it from a convert that investing a little time getting used to the proper way pays dividends for the rest of your life.

          2. People keep imagining that metric vs. customary is a matter of binary fractions (fractions where the denominator is a power of 2) vs. decimal fractions (fractions that are easily expressed numerically using a decimal point). This is NOT the case. I have never seen a gas pump that measures in 1/16ths of a gallon – all modern ones (at least) measure in gallons and tenths of gallons. I have never seen an odometer that measures in 1/8 miles. All the ones I’ve seen indicate whole miles and tenths of miles. I have never seen a CAD-produced circuit board drawing in 1/64ths of an inch – they’re specified in thousandths or hundreds of an inch. There are many, many things that are already measured using numerical values with decimal points, that are not metric. Please stop confusing the two things. The problem is not that we call things out in 1/16ths of an inch – those are very easy to convert; the problem is that we use inches at ALL.

          3. @TGT You cheated you compared adding meters and centimeters with adding millimeters and millimeters.
            Quick what is 1.270nm + 259.0pm, now what is 3/8m +5/8m.
            If you are measuring manufactured things you will likely find the dimensions to be nice multiples of the units the designer was using 4″ or 100mm doesn’t really matter when designing a coaster but most people would not design something to be 101.6mm without a good reason.
            You want weird at work we have 20mm threaded bolts with 15/16″ hex heads.

          4. @TGT, what is wrong with an inch, precisely, given, as you say, that it is frequently used with decimal division? How is a centimeter better? I grew up with metric and later moved to the U.S. I find conventional measure to be more convenient in almost everything I do except (and very rarely) when I have to convert measurements between disciplines (e.g. mechanical engineering and home carpentry or cooking and chemistry.) Then it just takes a bit of care to get conversion ratios correct.

            Standards are not a dividing issue because all conventional measure units are based on the same standards as SI, now, and have been for some time. One of the things which may have been forgotten if all you know is SI is that the units of conventional measure (and there are many of them) were decided based on their usefulness to the task to which they were applied as opposed to being nearly entirely arbitrary and so, within their domain of use, are generally much easier to use. The base 12 and base 2 divisions work really nicely for the kind of scaling done in the kitchen and in carpentry (we don’t often cut a recipe by 5 or 10, but often by 2, 3, or 4. And fractional measures in carpentry can be easily done in one’s head without resorting to a calculator. 1/2 of 5 3/8 is (2 1/2)+(3/16) is (2 8/16)+(3/16) is 2 11/16, for an example of what carpenters might do when finding the middle of a thing. For any but the simplest measures, the same computation using SI requires more complex mental arithmetic and might just require a calculator.

            All that said, the speed of light is one of the few true constants we can use for building standards. In so far as convenience is concerned, which is more convenient? The speed of light is about 300,000,000m/s or about 1,000,000,000ft/s (both to within 2% and both by accident.) I mean, is the distance from the north pole to the equator running through Paris really a useful standard (the original definition of the meter)? Of course, now both SI and conventional measure use the speed of light to define distance units, it just so happens that feet divide that distance, given our also arbitrary time units, more nicely than SI units.

          5. “There’s almost nothing salvageable about imperial units except the fact that a few people are very used to them.”

            You undercut your argument when you refer to the 3rd most populous nation on the planet as “a few.”

        2. The US is like that about literally everything.

          How can we provide health care? It’s impossible.
          How can we provide effective rail and public transportation? It’s impossible.
          How can we reduce gun violence? It’s impossible.
          How can we modernize broadband providers and break up abusive monopolies? Impossible.
          How can we stop sucking at measuring things? Impossible.
          Literally everything. We ignore the very obvious fact that every other part of the developed world has figured these things out and make up excuses of why we can’t. It’s ludicrous. I can’t believe we think a can-do attitude is part of American culture.

          1. “It’s ludicrous. I can’t believe we think a can-do attitude is part of American culture.”

            Well, we did manage to render 60 years of television standards useless a decade ago!

          2. 1. Get yourself some insurance
            2. Supply and demand where the demand is noisy enough to get noticed
            3. Stronger communities where people look out for each others mental health
            4. Prank call the bosses and otherwise torment them (Call round their house at approx 4am and ask)
            5. It’s not the units it’s how you use them
            For a country based on the idea of people doing things for themselves without asking the government to do things for them, you are pretty asky. If you don’t stop asking for things like this your government is only going to expand and you are less likely to be left alone to enjoy the right to life liberty and pursuit of happiness. I’m in the UK :(

          3. And yet, no other nation has sent men to the moon. Maybe the metric system is not that great an advantage.
            We sure don’t need advice from Euro-weenies on engineering.

          4. men on the moon? An impressive accomplishment, but while you’re basking in the glory of something your parents generation did, we’re enjoying healthcare, not dying of guns, not feeling the need to arm ourselves for defence, …

            The best thing America has managed for a generation is Trump’s meeting with NK.

          5. “And yet, no other nation has sent men to the moon.” maybe because they saw that it is useless an very expensive? money you can use to make heathcare work?

        3. If you want to foot the bill to both rewire ~126 million homes in the U.S., and re-equip all the factories making anything with a plug, AND start up a business making plug adapters for all the legacy equipment. Feel free. But make it snappy. I don’t want the changeover to take more than a couple months.

          1. First, I do not agree with the need to convert the U.S. to 240 V, or some other receptacle specification. Both of these would be pointless without also changing the grid to 50 Hz. Which would be a really bad idea. First of all, it would be unimaginably expensive, and second, the flicker from 50 Hz fluorescent lamps is much more noticeable than 60 Hz. This is one case where the U.S. got it right. 60 Hz also costs us far less in the size of transformers and capacitors needed.

            But second, conversion to a new standard of outlet in itself wouldn’t be that difficult – all you would have to do is replace the plugs on power strips with the new type of plug, to plug in a bunch of older devices. Also, I’ve seen pictures from places in Europe, which used to have several different standards for AC receptacles, and the solution was to mount new receptacles right next to the old ones. It’s not that big a deal.

          2. While we’re (you’re) at it, rewire for 400Hz, as it is more efficient (smaller transformers, smaller components to filter pulsating DC. Oh! Wait, Let’s (i.e. you) convert the US to a DC distribution system!

          3. A lot of this discussion ignores the fact that the US has largely focused on improving safety of the infrastructure rather than changing plugs. To name a few examples, tamper resistant outlets (which have shutters on the power pins) have been required by most jurisdictions since 2008, as have arc fault breakers. And of course GFCI have been required in some locations since the early 1970s.

            Shielding on the base of the blades (as Australia now requires) would be a great improvement, but it’s not like the US has done nothing. I’d much rather see a standard for residential low voltage DC than a ton of extra work on AC outlets.

        4. In literally decades of plugging and unplugging, with utter disregard for the momentarily exposed energized conductors, I have been shocked exactly zero times.
          That is not to say that inherently safe designs aren’t an improvement. It’s just that, in practice, this isn’t much of a problem.

          1. When I was pre-school, I stuck something metal into one of the slots in an electrical outlet.
            I learned that lesson well, I didn’t see a need to ever repeat that experiment.

          2. I recently read an article on how in the UK some places are making less heavily supervised less injury-proof playgrounds so kids have a chance to develop properly. It’s a great idea, taking away any chance of getting hurt is not a good thing IMHO as consequences teach responsibility. Our power is at 120V I have personally been zapped on a few occasions with no permanent damage. If we were pushing higher voltages I could see more reason for concern, but honestly… its fine. Kids need to get zapped when they do something stupid.

          3. @Derek Anderson – and how did your kids fare? I hope they weren’t seriously injured. It isn’t typical for 110 to cause any permanent injury but I suppose it is possible. I’ve personally been shocked many times working on live wiring (consider *actual* risk versus reward and the possibility that I had a boss who required OSHA violating practices and a job that I needed very much before you infer stupidity) and found it to be a mild reminder to be a bit more careful.

            I would guess they walked away with no more than a life lesson not to do what they did to get shocked, again.

          1. Shep: The fact (if it is a fact) that some dimensions come out to convenient numbers in inches, doesn’t excuse keeping inches around. I’ve seen plenty of chips with 1.27 mm pin spacing. These used to be chips with 50 mil pin spacing (0.050″), but we’ve gotten used to 1.27 mm. No need to use inches. And similarly, since 2×4 lumber isn’t even close to 2″ x 4″ in size, it would actually be more useful to call them 40 x 90 mm, which is a much closer approximation. Studs made from these and placed on 16″ centers can be placed on 400 mm centers, and nobody would know, since carpenters aren’t that precise anyway. Inches are just lazy.

            By the way, Canadian receptacles have the same dimensions as U.S. receptacles. The blades (non-polarized) are 0.25″ wide (6.35 mm) and 0.06″ (1.5 mm) thick, on 0.5″ centers. Polarized blades are 0.315″ (8 mm) wide. Canada also uses one side hot at 110 VAC and the other side neutral, with uninsulated blades on their plugs. Canadian power is indistinguishable from U.S. power, but I don’t hear anybody saying, “those stupid Canadians use dangerous power receptacles with non-metric dimensions”.

          2. @BrightBlueJim Stud size is based on a standard, which specifies than a nominal 2″x4″ rough sawn green board will be a 1.5″x3.5″ finished board. This confuses no one in a trade that regularly uses them. As 1/2 inch is not uncommon to see, calculating with those numbers, on the very rare occasion a framer will have to, is not an issue. Since your argument is that inches are harder to use (I don’t agree, but I’ll go with it for a moment) how, then, are inches, “lazy”? Also, Canada also uses 2x4s and other inch and foot dimensioned lumber.
            Also, how is that unit size (and division) convenience isn’t an excellent argument in favor of a measuring system? I think it is.
            I grew up in Canada with the metric system and later moved to the states. I find conventional measure with it’s convenient units and useful divisions, to be superior for most day-to-day activities, including machining, and SI to be superior in scientific endeavors where complicated unit conversions are much more likely to happen.
            Insofar as precision of the standards, there is no difference between the systems as they both use the same standards.

  1. Though it was explained to me long ago about the ground plug top advantages the lifetime of seeing all photos (like yours above) with ground plug down have permanently embedded in my brain the (in)correct way to install an outlet.

    1. Ground-pin-up also stops them from triggering our brains’ automatic face detection. That’s generally irrelevant to adults, but it’s been found to make small children less interested in playing with them.

    1. I remember getting schooled on this once when I was a loudmouth know-it-all teenager, was visiting my grandparents and Grandpa and I were visiting some friends of theirs who had just finished building their house, and I asked who the genius was who installed the outlets upside down. Rather than getting mad though, my grandfather and his friend explained to me why “upside down” was actually the right way.

    2. The incorrect orientation is so ubiquitous, many plugs have been designed around this. Look at any wall wart that has a ground connection – every one of these I’ve seen has the ground pin at the bottom when plugged in with the wire exiting at the bottom. Same with refrigerator plugs with right-angle wire exits. All of these that I’ve seen expect the ground to be at the bottom of the socket.

      1. True, that’s going to be a pain for a while. The way a grounded wall wart is weighted means it tends to fall out of the wall more easily if you plug it in to a ground-up outlet.

    1. The electricans I work with are aware of and using the “ground pin up” layout, when they work in one of my houses they flip the outlet they’re working on, but leave the rest as they aren’t getting paid to flip them all and the house was built before the code required it. That might be why you have both orientations.

    1. It’s like a solder-lug connection. If the cable’s too short, you wrap a bit of bare wire through that hole and stick the other end into a socket.

      I’m joking, but I’ve seen photos on the internet of people doing exactly that. What a world.

      1. You’re wrong! You should use a nut and bolt on that hole, you will have a much stronger connection that way! ;-)
        If people are worried about exposing mains connections, just use electrical tape.

        Disclaimer: don’t try this at home!

        In Europe (where I live by the way) you can just use screw terminals. People are always trying stupid things…

  2. The industrial one is much different because it’s designed for 20A, not 15A. That alone necessitates bigger conductors; this would be the same even for a consumer-grade product. Note the extra terminals at 90 degrees.

    1. The industrial one shown is not a 20A unit. The triple wipe is used across the better quality units, but, more to the point, the contact pressure is applied at several points over the blade, not just at the retaining detent point like in the consumer model. In 20A configuration, only one stab is lateral, so if that was the primary reason, it would only be the neutral (grounded current carrying) contact.

    2. If that face came off the industrial rated outlet, then Leviton uses the same innards for outlets to be used for higher current and/or voltage, with one or both blades rotated 90 degrees. Put a different face on and charge more for essentially exactly the same product.

      With two horizontal blades it’s for 240~250 volt only. With the ground hole on top, the right slot may be horizontal or sideways T shape for 115~120 volt 20 amp, with the T slot allowing standard 15 amp plugs.

      Also note the ground hole is NOT round but a D shape. That allows both round tubular or solid ground pins and U shaped pins that are made from stamped sheet metal.

      Then there’s the “Despard” style that packs two or three 15 amp (grounded or ungrounded) outlets and/or switches, rotated 90 degrees, into a single standard size box. The type with one switch and one outlet use the switch to operate the outlet. No guessing as to what the switch controls.

      Back when what would become the NEMA standard was coming into use, there were four slot outlets with them all horizontal. The upper and lower for the new parallel blade 15A and two inline in the middle for the old Hubbell inline blades.

      What I find interesting is many outlets used to be treated as somewhat of a decorative feature, with various designs molded into their plastic. At some point the manufacturers quite that, making them all smooth blandness.

      https://www.plugsocketmuseum.nl/NorthAm1combi.html

      https://www.kyleswitchplates.com/identify-an-old-switch-or-wall-plate-cover-find-replacement/

  3. The “ground on top” thing is fairly new, and makes very little sense.
    The push-in backwire on residential outlets scares me much more – I doubt they can handle full current..
    Residential outlets are now supposed to have “shutters” to prevent kids from pushing stuff into the slots…from my limited experience, these seem to work about as well as you’d expect…which is to say, not at all.
    The outlets installed in my house are “cheap residential grade” and barely retain the plugs. I’m in the process of changing them to industrial grade.

    1. The idea of the ground on top would be if one were to drop something conductive and metallic with the plug pulled slightly out, it would tend to shunt to ground and not short out across the hot and neutral. It’s not really a solution to the fundamental problem of a poorly designed socket though it is arguably a slight improvement from the “standard” orientation.

      1. You’re close…the idea is that the object hits the ground pin and hopefully falls away without causing a short; alternatively if it doesn’t fall away, the object is resting on the ground pin and not the hot blade. Shorting to neutral or ground doesn’t matter, they are both connected together at the breaker panel.

        1. True. Shorting ground to neutral shouldn’t (generally) be an issue. But you can still short hot to ground quite easily.

          So it depends on what you drop on it exactly. It could be deflected completely, it could hit ground and stay there, it could short neutral to ground or short hot to ground.

          It’s certainly better than falling across hot and ground and almost always becoming part of that circuit though.

          Somewhat more relevant for GFCI circuits as well. Ground is typically for the safety of the user. It’s not really required for most typical circuits to function.

    1. I lived in a house built in 1958, which was a model home for a southern California tract development. For the light switches it had momentary buttons a little smaller than the actuator on a Decora switch, and mounted in a standard electrical box. These were all wired to a relay chassis in the attic, and used 20 VAC signalling, with all of the light fixtures wired from there. The relays were ratcheting rotary relays, which made the wall switches alternate-action. The big selling point was a panel in the master bedroom that could turn on or off six circuits, including the front porch, back porch, living room, hallway, and kitchen.

      Huh. It looks like these are still made, under the brand “Touch Plate”.

      1. I bought the house across the street to use as a rental and it has Touch Plate lighting. It works well. And, yes, the company is still in business, offering a digital version for new installations or upgrades. They were pioneers in a type of home automation.

      2. A friend lived in an older home that had the exact same system for a year while in college. All the switches were worn out so you had to hit the button a couple times for the light to come on. The bedroom panel was handy but I could have lived without it. In one of the bedrooms, someone had replaced a broken pushbutton switch with a normal light switch. When you flipped that switch on, the relay for that circuit would hold in place and not allow any other switches in the house to toggle. When she first moved in, there were a couple nights where they had to unscrew lightbulbs from the ceiling because the bedroom lights wouldn’t turn off. After calling the landlord and having an electrician come out who had never seen that kind of low voltage light switch system before, we finally narrowed it down to that normal-style switch and pulled out.

        The system seems like a neat idea and with the high price of copper in the early 70s, I’m sure it made fiscal sense but the added complexity really just made things more of a headache when something breaks. Trying to figure out which relay goes to which room was impossible since they were all unlabelled and they would click so hard that you couldn’t tell which one had the strongest vibration. I’m pretty sure the original owner also had the built-in 8-track and turntable installed in the bedroom as well. The house was wired for sound but the speaker wire in the walls had long since oxidized and become unsuitable for music. I’m surprised there was no intercom system installed. All the rooms had dual phone jacks so I wonder if the second phone line was used as the wiring for an intercom system since having that many second lines around the house doesn’t make sense for voice or dial-up and phones that support two lines don’t need two cables.

  4. It is interesting to compare US plugs and sockets with UK plugs and sockets. In general, US ones are a marvel in making something effective at low cost and the UK ones are bombproof but large, heavy and expensive. There is a lot of attention to safety on the UK ones, probably because line voltage is 240v. The plugs are individually fused, the upper part of the blades are nowadays insulated so that no live metal is exposed when connecting and disconnecting. The pins are massive rectangular machined pins capable of carrying much more than the rated 15 amps (fused at 13 amps or less to give a 2 amp margin). I expect a group of engineers designed the safest possible plug without regard for cost. In the US, cost usually seems to be the prime factor. This can also be seen in the simplicity of RJ11/RJ45 connectors. I wonder if this approach could be part of the reason for a lot of the commercial success of the USA. I used to be a database consultant and initially couldn’t understand how relational databases with their limited capabilities could be successful when there were much more capable alternatives. Of course the answer is that “Good enough is good enough” and cost and/or simplicity are just as important as capability.

    There also seem to be two approaches to hacking. One is to make something that works quickly, cheaply and simply. The other is to lovingly over-engineer something much better than is needed just for the pleasure of building something really good. I find myself drifting toward “perfection” but pulling back somewhat to make sure I actually get things built!

    1. Cost and/or simplicity are not always just as important as capability. As usual, it depends on the application. Do you want your rocket to be cheap and simple or well built (but not overly complicated) and cost more but be more likely to not explode or be able to complete the 5+ year mission successfully?

      Some people drive their cars into the ground because they just don’t care but they also can easily get a tow.

      Sometimes things have no practical purpose. Building a miniature V12 engine from scratch just because you can is probably not very useful or practical but can be a learning experience or retirement project. Is that a project for a paying client or something done for personal enjoyment? What about reusing a project that isn’t quite right but might do a good enough job if modified? How much should you modify it?

      Also, sometimes when you are iterating, just try things out and move on. That makes sense if you are going to rapidly change things up anyway. But that too is not always the best approach. It truly just depends on the specific goals for the project and each project is different.

          1. I stepped on a 27C1001 32pin DIP EPROM as a kid, it hurt surprisingly little.
            But had a neat row of holes in my heel for days, and it took force to pry it out. *shudder*

          2. Obviously a man who has never stepped on a British mains plug. I once heard of a company who shipped an excess lot out to Canada to sell to bear trappers.

    2. American wiring uses much more copper than British, because of the lower voltage, but 120VAC is less dangerous and way less painful than 240VAC (speaking as a person who’s been hit by both plenty of times). British outlets and plugs compensate for the higher voltage by being safer at the user interface.

    3. The UK plugs are over-engineered because of the old habit to wire houses with ring mains.

      There are certain fault conditions where you may get 64 Amps through the socket, which means it has to be built like a brick shithouse to make sure it’s the fuse in the plug that pops instead of causing an arc in the socket that burns down the house.

      1. Ring circuits were used to save copper, on the theory that wiring with half the diameter could carry as much power if it were run around in a loop.

        It works, when all that’s connected are low power things like lights and radios, fans, and those newfangled vacuum cleaners. But then along came the electric heaters (or electric fires as some of you Brits are wont to say), electric stoves, ovens, clothes dryers, water heaters etc. A ring circuit still works as long as the high power equipment is near the middle of the circuit. But put it closer to one end and the resistance imbalance will cause most of the power to try going though one side of the ring. Oops. Your wall’s on fire,

        1. Not half the diameter, but half the area. Same idea though, and the unbalanced load concern also made it necessary to specify 2/3 the cross-section, meaning you save about a third of the copper.

          With two paths for current, reactive loads like motors can cause interesting current patterns that prevent the main fuses from tripping, and the voltage between sockets can vary considerably depending on the load condition.

          These problems with weird connections is also the reason why british appliances come with cables that can be used for mooring the Titanic – because the sockets can provide so much current without any breakers flipping, it’s the appliance cord that can catch fire just as well.

          In the EU all the high power equipment typically have their own dedicated three-phase wiring. Cooking stoves, water heaters etc. have their own breakers at the box, which also means you get nice things like high power induction stoves that heat up just as fast as gas.

          1. British appliance cables are protected by the fuse in the plug. There’s no way massively excessive currents could get to the cable without the fuse blowing.

  5. The comparison between the Leviton 5320 and 5252 is a little mis-leading because for the 5252, Leviton uses the same 20A duplex base for both the 15 and 20 amp version duplex outlets. The only difference is the thermoplastic face plate (the 20A has vertical and horizontal openings for the neutral blade on 20A cords). There is no purpose for a triple wipe blade on the 15A plug as the horizontal receptacle is never used. Being that a 20A base is being used for a 15A outlet, of course it looks much better than the residential 15A-only 5320 outlet.

  6. You say 15 inches up. I don’t know if that is code or not, but when I put outlets in what is now my shop/lab/studio, I put them all in about 44 inches up from the floor, which puts them about 12 inches above my counter tops. I have thanked myself ever since (but who knows if this is a code violation). I also wired them all ground pin up and think that is a good thing. It lets me be a know-it-all when people complain about it. (I tell them, “well you know that is code don’t you?”) I am used to it and I don’t truly care one way or the other.

    I also put in 4 outlet boxes and have forever thanked myself for that also. Should have done more of that.

    Nice article. I like this sort of thing.

      1. Yes !! Another advantage I hadn’t thought of !

        Mostly for me it is not crawling around on my hands and knees bumping my head and saying harsh things.

        Every outlet should have 4 sockets, that should be code.

        1. Would those be the same shops and garages, where welding and cutting torched may be in use? Not to mention big fat sparks from engine starting motors. These days ADA construction guidelines probably dictate the height of light switches and power cord receptacles.

        2. It’s nice how the girl, despite being an accomplished hacker and technical journalist, despite not having even written this article, gets complicated terms like “gasoline” explained to her by the gallant readership. Other British, or Australian or wherever else commentors and or article writers get nowt.

          Without presuming to speak for her, speaking as an English person myself (though a male one, and therefore of course expected to know these things), Jenny probably already knows that Americans call petrol gasoline. We get quite a few American films and TV programmes over here, and have done for a few years now, since Hollywood whizzkids figured out how to send 35mm over 240 volts. She probably also knows how to read “aluminium” and “colour” spelled both ways.

          I know she’s a girl who owns more than 2 screwdrivers and therefore we all fancy her, but it’s getting embarassing. Perhaps we should we all start emailing explanations of Ohm’s Law to Jeri Ellsworth in case she giddily forgot, and give Jenny a rest to eat chocolates with her curlers in for a couple of weeks.

    1. I don’t think code specifies a distance. I have heard that the standard location is one hammer length above the floor, because that’s the only tool the guy nailing the boxes to the studs has with him.

      1. Yeah, why would an electrician carry a tape measure? I think it’s more likely the other way around: electricians, on discovering that they left the tape measure back in the truck, figured that 15″ was about a hammer handle plus a little bit.

      2. When talking about new wiring in large housing developments, “Guy who hammers the boxes to the studs” is more accurate than “Electrician”. There may be a licensed electrician involved in the process, but most of the work is done by cheaper labor.

    2. Every new construction garage i’ve seen has them around that height, usually in my areas its because the lower portion of the garage wall is a foundation wall, so its concrete to about 3 feet up.

      Is there any code for the height of an outlet? I’ve seen them in some crazy places for older security cams and televisions. If you check 3 houses you’ll probably find 3 different outlet heights. When I started my first low voltage job I asked an electrician what the code was for outlet height. He held his hammer up to the wall, marked where the top of the hammer landed on the wall, and said “bottom of the box goes there”.

      1. They may in fact even be in the floor as long as *local* codes allow it (I don’t remember if they require a protective plate). ADA has recommendations for mounting heights for both normal convenience receptacles and light switches, but those aren’t usually applicable outside of commercial work. Still a good guideline.

        “Hammer handle” height is pretty common, as mentioned usually the guy nailing up boxes and drilling holes is *not* the electrician.

  7. I’m a little disappointed that the article didn’t cover the more recently change to polarized plugs, which you can see in the Levitton plugs, where the neutral blade is larger than the positive blade. This is to make sure you have the plugs oriented correctly, but I can’t recall the details off the top of my head.

    It would have been also good if the article discussed some of the designs which didn’t survive and make it into service. And also how did Hubbell get all the appliance makers to put on two pin standard plugs? And when did three prong become required? Lots of history left out here, all of which relates to this socket.

    John

      1. Not so much of a life saver, since so many outlets in the US are mis-wired. Even today, I find mis-wired outlets all the time. Even worse, though, in a YouTube thread about cheap CNC engravers, we discovered that some 3-wire power cords are even miswired, with the blue wire is hot and the brown neutral.

        Never, EVER design something depending on the correct wiring of the mains connection for safety. Transformerless TVs and radios were designed with no metal parts accessible outside the cabinet that connected to the chassis, but people often replaced broken all-plastic knobs with knobs using setscrews, introducing a shock hazard. There doesn’t seem to be any way to make things truly fool proof.

      2. when most radios and TVs had metal chassis tied to the neutral, both pins were the same size. The wider plug is a recent innovation (I remember when they first started showing up, and wouldn’t fit into many outlets)

        1. Yes, that’s true, but again, once they did this to avoid the shock hazard to people using pliers in place of the broken knobs, the assumption was that when people replaced outlets with the new polarized and grounded type would wire them properly. Not a valid assumption.

          1. I was thinking about this comment this morning as I plugged my Black and Decker juicer’s NEMA wide bladed two pronged plug (no ground, one of the two blades has a wider head) into a NEMA power outlet with different width blade sockets and a ground socket.

            Only, I didn’t pay attention and I put it in backwards. It fit just fine. Worked just fine.

            Why is it possible to put a polarized plug in backwards? What’s the point if the device physically fits the wrong way anyway?

            This is not a no-name device, and I installed quality power outlets a few years back.

    1. That change pre-dates the ground pin…

      I’ve lived in several houses old enough to have the two prong outlets that you had to get the 2 to 3 prong adapter for.

  8. It seems to me that a history of the USA style outlet/plug should include the addition of “polarity” and those childproof shutters. These backwards compatible changes achieved some of the safety advantages of the UK plug without adding bulk or cost to the plug.

    1. I think if you’re really bothered about the cost of a bit of extra copper, why not just do a 50% shrunk version of the British plug? It’s clearly the best mains plug there is. They thought of everything when they designed it. Though it did replace earlier designs, including ones with round pins, so there will have been accident reports to guide what features were needed.

      I’d say just start using the British plug, but then with international trade, and China’s quality control standards, there’d be a fair risk of stuff ending up on the wrong side of the Atlantic and ending up exploding from too high a voltage.

  9. I’m glad you decided to talk about this… I think America should give up on the 120 and go 220-240 v for everything! It would simplify our electrical boxes and give more power to our Outlets… Without changing any wiring…. We would only need two wires coming into the house instead of three, more savings…

    1. “We would only need two wires coming into the house instead of three, more savings…”

      Err? Most homes in America are not powered by 3 wires being fed into the house.

      1. Two hot and a neutral. I suppose you could claim that the neutral isn’t powering the house, but it’s essential if you’re going to have both 120 and 240 volts available without having a large transformer in the house.

        1. You may recall, the neutral is for grounding of errant currant.
          On our 66 megawatt turbine, the ground leg needed a large resistor bank just above the ground to prevent explosions if or when a short went to ground.

          1. Even without fault current, split-phase service has to be designed with the assumption that it’s a normal case (i.e., not caused by appliance failure) where one side will be carrying all of the current. So for example, a house with 100 A service may at any time have 50 A of load on one side and zero on the other, meaning the neutral has to carry all of those 50 A back to the transformer. This may not be due to bad design, but just by random variation in the combination of loads being used at any given moment. It is only likely to happen if most of the household loads are 120 V appliances or lights, but this isn’t all that unlikely. Even with high-power loads served by 240 V circuits, there are valid cases where almost all of the load is on one side. For example, some appliances will feed a heating element with 120 V under some circumstances and 240 V for others. I haven’t seen any new appliances doing this, but there were a lot of appliances (many still in service) from the 1970s that did this.

            So the bottom line is, all three conductors feeding a split-phase house have to be the same size.

  10. The correct way of orienting grounded outlets helps support grounded plugs better from pulling out. Gravity has to pull out on that bigger pin first, which often has a hole at it’s base making it easy to break off.

    Wipe that smiley face off that EQ!, a sure sign of incompetent use. Outlets don’t need it either.

  11. “with three holes arranged in a way that can’t help but stimulate the facial recognition firmware of our mammalian brain”
    My Dad and Grandfather always referred to it as either being surprised ghost or vampire cyclops depending on the orientation.

  12. What I want to know is, who’s the idiot who standardized flat blade screws for the faceplates (hmm, which shape would be perfect to accidentally electrocute yourself?) while they could have chosen Philips and be impossible to stick in anything but maybe ground?

      1. Meh, you’re not going to strip a head on an outlet plate cover, they’re zero-torque. Philips is self-centering, so tends not to slip. Also, the new-fangled hammer drivers pretty much fix the cam-out problem where torque is needed. Biggest problem now is pushing hard enough that the 2000 RPM driver doesn’t suck the screw into the work faster than your arms can accelerate the driver in your hands. Breaking screw heads off faster than you can release the trigger is also a problem.

        Of course, you could say it’s a Darwin Award awaiting the fool who doesn’t throw the breaker first, but sometimes people put the breaker number on the backside of the plate so it has to come off beforehand, enter other reasons people will take risks… like using power tools that need an outlet and all the outlets on the floor share a breaker.

          1. They’re definitely an improvement over Phillips for anything that needs to handle some torque such as wood screws.
            Though the reason for slotted screws on faceplates is probably aesthetics it just looks better.

        1. The problem is rather that people -try- to tighten them because they think they must, push really hard and then strip the head of the screw when it cams out.

          And they’re using the wrong screwdriver anyhow, typically one that’s much bigger than what really fits the screw.

  13. For all of the 220V lovers, one of the things that’s been lost to the sands of time (fortunately) is the older British system of having different outlets for different service amperage (this may have been BS 546, but I’m not sure). My childhood in an American expat family in the early 1960s was one of many, many different adapters in an old “semi-detached” house.

    1. Here in the US, you will find different sockets/plugs for 220/480 single or 3 phase outlets. They are varied to respect the maximum amount of current to be supplied to a particular device. IOW, a 50 amp outlet won’t allow a 30 amp plug.

      1. Seems rather backwards. A 30A load in a 50A socket should be no problem, same as a 0.5A load in a 15A socket is fine.

        In AU, we have 10A and 15A single-phase sockets; the 15A have a wider ground blade so you can plug a 10A load into a 15A socket but not the converse.

        So many annoyingly incompatible plug options for 3 phase (e.g. with/without neutral) but it’s not a huge issue because most large machinery is used in just one place – the plug is just a handy disconnect, not so much a means of portability.

        1. This – the notion that it’s okay to plug 0.5 A devices into 15 A circuits – actually doesn’t make sense. The purpose of fuses and circuit breakers is to protect the house from being set afire as a result of electrical failures. If you plug a 0.5 A device that uses a 24 AWG power cord into a 15 A circuit, and you have a fault in the device that causes it to draw 12 A, there’s a good chance that the power cord itself can become an ignition source where it passes over carpets or other combustible flooring, but zero chance it’s going to trip the circuit breaker.

          Look at how circuits are fused in automobiles: the fuse size isn’t based on the current demand of the load; it’s based on the size of wire that connects to that load. This way it doesn’t matter if the circuit fails as a dead short, or as an arbitrary overcurrent condition. As long as the fuse will blow before the wire’s insulation melts, a fault won’t start a fire. This is how it should be in mains-powered electrical devices, and this is where Great Britain got it right: you put a fuse in the plug, which is sized according to the current handling capacity of that cord. This should be required in the U.S.

          1. If you plug a cheap USB charger into the car and it melts itself without tripping the outlet’s fuse, does that mean the car’s fuse didn’t do its job? The circuit breaker is there to protect the wiring between the service panel and the outlet, just like the fuse on a car circuit.

            That hypothetical device should have its own fuse if it has a fault state where it can suddenly draw 24 times its normal current. That’s a failure of the device, not the infrastructure.

          2. If you plug a faulty device into the cigar lighter receptacle in an automobile, even if the device turns into slag, the receptacle (and the car) should survive. I mean, it’s designed to have a cigar lighter plugged into it!
            Also, most devices that plug into a cigar lighter receptacle that can’t handle 30 A include a fuse in the plug, because their manufacturers don’t want to buy you a new car.
            I’m not sure how dollar stores protect themselves against liability for damages caused by faulty merchandise. I don’t recall seeing a disclaimer posted in the stores.

          3. “That’s a failure of the device, not the infrastructure.” Good point. However, in the U.S. we don’t seem to have an authority for devices that plug into mains power. For infrastructure, we have the National Electrical Code, but for devices, the best we can manage is a commercial entity (Underwriter’s Laboratories) that tests devices, but no legal requirement that devices sold in the country be tested by them or anybody else.

          4. In the UK, circuit breakers protect the property wiring and the plug fuse protects the appliance wiring. As safety is a priority in the UK, the electrical code now also requires RCDs (GFCIs) on new domestic wiring. They also require professional inspection for a lot of DIY wiring.

          5. Bull.

            Your house’s breaker or fuse is there to protect your house wiring from having too much current drawn through it and setting your walls on fire. Your devices should have their own fuses rated properly to prevent themselves from catching fire. A separate outlet with wires running to a separate fuse for every amperage of device you might want to plug into the wall? Ridiculous!!!

  14. I have often shown and suggested to homeowners, a “tool” I “make.”
    Take a 3 wire polarity tester, add a 3 to 2 adaptor and plug that into a 2 wire to lamp-bulb adaptor. Now a homeowner can check polarities of everything, 120vac, to include lamps and lights, which I find to be the most often incorrect. If the polarity indicator has the button switch, GFI’s are also testable. One of my all-time best “hacks.”

  15. “For reasons unknown, though, Hubbell altered his design in 1912. The two blades were no longer in a line; each blade was twisted 90° to form the familiar parallel arrangement we see to this day.” While I have no supporting evidence to offer – cracking of the front plate. The slots and screw holes all lined up in a row, turn the slots 90° and it becomes much more durable.

    1. It also would reduce the chances of touching one of the conductors while removing it since the they would not be as close to the edge of the plug.
      Maybe the first safety feature to be added?

        1. While many things are done weird in the us, switched outlets are not. The switch is near the door for example th uk/aus thing i never understood though. Why on the plate?!

  16. Reminds me of a bit of automotive trivia I heard: The one part of your car that is practically identical to one found on a Model T is the Schrader inflation valve on the tires.

    1. Of course, if you want your audio quality to be top-notch, your AC plugs MUST be gold-plated silver, using oxygen-free copper wire. How else can you expect to get decent transient response and phase linearity? I know _I_ can hear the difference…

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

w

Connecting to %s