Design for a universal plug


I've written before about both the desire for universal dc power and more simply universal laptop power at meeting room desks. This week saw the announcement that all the companies selling cell phones in Europe will standardize on a single charging connector, based on micro-USB. (A large number of devices today use the now deprecated Mini-USB plug, and it was close to becoming a standard by default.) As most devices are including a USB plug for data, this is not a big leap, though it turned out a number of devices would not charge from other people's chargers, either from stupidity or malice. (My Motorola RAZR will not charge from a generic USB charger or even an ordinary PC. It needs a special charger with the data pins shorted, or if it plugs into a PC, it insists on a dialog with the Motorola phone tools driver before it will accept a charge. Many suspect this was to just sell chargers and the software.) The new agreement is essentially just a vow to make sure everybody's chargers work with everybody's devices. It's actually a win for the vendors who can now not bother to ship a charger with the phone, presuming you have one or will buy one. It is not required they have the plug -- supplying an adapter is sufficient, as Apple is likely to do. Mp3 player vendors have not yet signed on.

USB isn't a great choice since it only delivers 500ma at 5 volts officially, though many devices are putting 1 amp through it. That's not enough to quickly charge or even power some devices. USB 3.0 officially raised the limit to 900ma, or 4.5 watts.

USB is a data connector with some power provided which has been suborned for charging and power. What about a design for a universal plug aimed at doing power, with data being the secondary goal? Not that it would suck at data, since it's now pretty easy to feed a gigabit over 2 twisted pairs with cheap circuits. Let's look at the constraints

Smart Power

The world's new power connector should be smart. It should offer 5 volts at low current to start, to power the electronics that will negotiate how much voltage and current will actually go through the connector. It should also support dumb plugs, which offer only a resistance value on the data pins, with each resistance value specifying a commonly used voltage and current level.

Real current would never flow until connection (and ground if needed) has been assured. As such, there is minimal risk of arcing or electric shock through the plug. The source can offer the sorts of power it can deliver (AC, DC, what voltages, what currents) and the sink (power using device) can pick what it wants from that menu. Sinks should be liberal in what they take though (as they all have become of late) so they can be plugged into existing dumb outlets through simple adapters.

Style of pins

We want low current plugs to be small, and heavy current plugs to be big. I suggest a triangular pin shape, something like what is shown here. In this design, two main pins can only go in one way. The lower triangle is an optional ground -- but see notes on grounding below. In this design, a small low current plug can still go into a large, high-current jack. It is safe because it will tell the source not to offer that much current. The spacing of the pins is an interesting question. For example, a high-current plug might not have the triangles come to a full point, or it might put insulator at the point to keep the shape. In this case, the conducting parts of the pins can be spaced as far apart as desired to eliminate any risk of arcing. It may be designed so that the sockets also have different depths so that we can control what segments of the triangle make contact. The main goal here is that you can have a very small plug for a low power device and it is compatible with all sizes of larger plug.

Other designs are possible of course, such as blades that can just become longer. On the diagram, a few data pins are shown but there are a lot of options there. Another option is a system that simply has multiple pins (progressively larger) and the bigger plugs can use more of them to plug in.


Traditional high-current plugs have a locking mechanism. It is important to now allow disconnect when powered. A smart power system can have a shorter pin that detects if the plug is coming out, and shut off power before it can be disconnected completely. This may eliminate the need for locking for safety. However, sometimes locking is useful just to assure power is not interrupted by accident. The most common designs use twist-lock, which involves rounded blades which can turn in the socket. This does not allow the triangular design, but could support designs with multiple pins or longer pins.

Apple, with its magsafe connector, introduced new thinking to how tightly connectors should bind. When the cord leads to a fragile device like a laptop sitting on a table, you don't want the cable to lock or even resist being pulled out very hard. If somebody trips on a cable, you would much rather switch to battery for a moment than pull the computer off the table or damage the power cord. Apple used a magnet for this design and has patents in the area but the goals they aimed for are very worthwhile. It may not be possible to meet all the locking (and light-locking) goals we have with one connector.

We want connectors that don't come out casually, come out easily when we want to disconnect them, but which also come out when too much force is applied by accident. A tall order.


Grounds exist both for safety and to shield noise. Today's smart power circuits which can detect any ground fault and interrupt a circuit quickly may actually eliminate the need for safety grounds in new devices, but that will take quite a change in thinking. Because shield grounds like to wrap all the way around the connector, it is much harder to design a universal plug that can grow in size. It may make sense to put a shield ground only around the data pins, and put them in a row so that the connector can grow and increase the number of data pins. A slot connector (as is commonly found on newer data connectors like HDMI, micro-USB and SATA) can be designed to be able to grow and add more pins, allowing both new longer connectors to go into older smaller slots if those slots left extra space to one side, and for older plugs to go into newer jacks (that's usually much easier to arrange.) The main power pins would not need noise shields unless they will be used to send high frequency AC. This is not out of the question, as it is a handy way to send power for some types of lighting, or the intermediate stage power of a switching power supply.

Data pins

In general, as new data buses arise, they sometimes change the nature of their pins but often just want more pins. Dual-link DVI is just DVI with twice as many data wires, for example. While it can't last forever, the ability to add data pins to a data bus has value to keep a connector alive for longer. This might be particularly true in video cables. In the future, video displays may come with frame buffers, which would allow higher-resolution displays to be driven by older, slower video sources using protocols other than raw bit streaming. Connector compatibility would be nice. We managed to go from 10baseT to gigE on the 8 pin modular plug.

What voltage?

While smart power lines should be able to deliver a range of voltages and current limits, there is still some debate about what the most common voltages might be. There is a move afoot to power all PCs with pure 12 volts, for example, and generate other voltages needed internally with buck converters, which are now cheap and efficient. Oddly, at the same time, cars, which originated the 12 volt standard, want to move internally to a higher voltage.

The higher the voltage, the more power you can deliver over the same wire at the same current. The only reason you don't want to go as high as you can is that high voltages can be dangerous to humans, and will arc over a larger gap and need more insulation. Lower voltages like 12v are quite safe but need very thick wire to send any power. It is, however, easier to make batteries to deliver the lower voltages.

Many systems, including most phone CO equipment use 48 volts. Electric cars and golf carts all use more voltage. Home solar systems, even ones that use 12v batteries, tend to combine them to work at 24 or 48 volts. Generally a bit over 48 volts is considered the danger zone. You can feel a shock from a 48 volt circuit but it is unlikely to cause lasting damage, and it's hard, but not impossible for it to kill you.

Most other efforts call for DC. AC's main advantage is you can convert AC voltages up and down easily with older technology. This advantage is fading. It does have a few other advantages -- no need for polarization of plugs, and some devices will synchronize off its frequency. Its disadvantages are also strong -- it generates noise (the hum we all know) and flicker in many types of lighting.

Is there one type of plug that can do it all? Perhaps not. But as we go into the future and design new ones, we should learn from how old plugs became obselete and try to avoid both the old mistakes and new ones we can think up.


All your ideas are well and good. What about needing to be compatible with the USB cables that connect the phones to computers?

I don't want to have two connectors on the phone. And I want to be able to re-use the mini-usb cables I already have to connect to cameras, peripherals and what not.
And I want to be able to plug the phone into my computer to charge with those cables, not have to keep track of another funky cable.

Well, you are going to need to switch to micro-usb anyway. But USB really isn't enough power for a lot of applications. It isn't able to power a laptop or even charge many digital cameras at any decent rate of speed. It certainly isn't able to power my digital camera. USB has become a power standard by accident, because it was there. I'm talking about what we might do from scratch.

Don't overlook grounding and fire safety issues. Fire safety is particularly important in consumer devices, since they tend to be abused and poorly maintained while being used in fire risk locations. Heat loss goes as the square of current, so those people pushing 1 amp down the USB wires are loading those wires at four times the safe heat loss level. They are small wires. They are easily kinked and damaged. They are likely to be mistreated and ignored until they completely fail. The amperage limit was chosen as level where that failure would not involve enough heat release to both melt through the insulation and start a fire.

Grounding is much less likely to be a safety issue in the home or office environment at the power and voltage levels of this equipment.

The mini-USB choice is not bad for the loads of modern electronics in handheld devices. Look for something different (like Apple's connector) for larger devices. They need the higher power, but they have the larger size needed for higher capacity power electronics and connectors. I would push for either 36v or 44v nominal so that you can take advantage of the semi-conductors that are designed for new automotive or avionics use. These two standards are actually very similar from a semiconductor perspective. The automotive power standard tolerates a huge swing in voltage and power, in recognition of the automotive environment. The avionics requires much tighter power regulation.

The selection of a voltage is also affected by the natural chemical characteristics of components. If you have batteries involved, then you need them to generate appropriate voltages. The 6v, 12v, 36v, and 44v DC standards were chosen in part because the lead-acid battery cell is a 2v source. The voltages chosen are an integral number of cells in series in the battery. There are similar natural voltage levels that are optimal for the characteristics of the power semiconductors involved. Rather than re-invent all that, piggyback on the new automotive and avionics levels. Picking the automotive standard would greatly simplify automotive power for laptops. Lots of salesmen and families would like to be able to plug games and computers into their automobile.

One problem with triangular pins is the sharp edges and correspondingly tight acute angles in the socket. These are hard to manufacture and require tight tolerances in both plug and socket. With rectangular pins there's lots of room for error in pin spacing, and the round pin sockets are often set up internally to allow horizontal slop. Give or take a millimetre on the long side, less on the short side. So perhaps something similar to the existing UK inverted-T design would be better. That design is big but also carries 15A without stress, and the plugs are currently often fused so there is room to put in all sorts of smart stuff instead.

Alternatively, using some kind of ZIF technology at the high end would allow more flexibility - the 200W socket by your bed or in the conference room could be push-to-seat while the 20A one for your stove could have a locking lever that pushes the contacts onto the pins.

Audio jack style coaxial data pins might make more sense. Especially since most mains plugs now have plastic sleeving at the base of the pins to prevent finger-wrap shocks. Adding a couple more contacts for data would be easy enough. That would give you four or six connectors. Adding a slot with 4-8 more pins in the centre of the power plug would give you both more data, removal sensing and a low-power easy-pull socket for laptops.

The standard voltage should be less than 36V or it will need law changes in much of the world, as the minimal-laws voltage is set to that to allow 24v lead acid battery charging without needing licensed electricians to do the wiring. But today there's no real reason why you need more volts to charge - assuming a buck converter would allow a 30V or 32V standard with a bit of leeway for errors.

FWIW, frame buffering screens can't come fast enough for me. USB video adapters already do that, I'm just waiting for a monitor with one built in. And needing dual-link DVI to display text on a 30" monitor is pretty ridiculous.

I had not considered what you suggest about what you need an electrician for. But you can't put a lot of power down a wire at 24 volts. Even a computer that may draw 400 watts (not that it should) is going to require the sort of wire we use today, and hair dryers, vacuums, microwaves etc. are not practical. At 48 volts we can pull it off.

The triangles could be rounded at the ends, that's not so difficult. The goal with the triangle was to make the conductor thicker the higher the current rating, so you don't have to get very big before you can deliver even 100 amps. My goal, perhaps not attainable, is one standard that could work over the whole range, rather than having so many types of plugs. As noted, if plugs deliver smart power -- the source does not offer the real voltage and current until contact is assured, and cuts it if the plug is being pulled out or if there is a ground fault, then locks and safety features are not as necessary.

One option that could work is making the slots big enough for a decent conductor and not requiring complete fill. But there's a lot of history behind why power plugs are that shape, so more research is needed. Also, keep in mind that high current applications are pretty specific for the most part as generally higher voltages are used instead. Even electric stoves are often two phase to get extra voltage, or in industry 1500V or 3kV machines are used (I've serviced an 11kV/1MW motor... only 100A).

Planning to pump even 20A through a household appliance is a bit nuts. Even a 2.5mm^2 cable drops a significant fraction of a volt every metre at that current, but that size cable is not very flexible (think worksite extension cord rather than USB cable). So a 2m long cable with 20A might be 24V at the supply end, but you're going to get 22V or so at the receiving end by the time you've gone through the cable twice plus couple of sets of contacts (plug and fuse).

Also, contact area becomes very important at high currents and the odd current distribution in a triangular pin will be problematic. Three phase plugs here use round pins to get more contact area (they use cylindrical section contacts in the sockets), but with triangular pins you'll get less current in the sharp end but the same contact area, making for heating problems at high currents.

Then the triangle is out, and the extending blade becomes the likely choice.

The next question is polarization. It is obviously easy to design a plug to be polarized (the two slanted blades of Australia and a few other places are one obvious way) but the question is, do we want or need this? With smart power, you don't apply power until you have had some sort of dialog with the sink, except to give it some power with which to do this dialog, and it can use a simple rectifier to take either polarity.

Unpolarized systems have the advantage of convenience, you don't have to look at the plug. They are easier to plug in around corners or in dark places, you can put wall-warts in either direction, though in the smart power world you hopefully don't have transformers that plug into the wall, but you may have devices that plug into the wall without cords.

With GFCI you don't need a high-current ground conductor.

One of the goals here is to make things small, and have the same system around the world. Sockets/powerpoints are absolutely huge in some countries now. There is a nice aesthetic in being able to make low-wattage sockets be small and unobtrusive.

It may be the right design is just two parallel blades, and the blades can get longer if they want to carry more current. If the blades are along the same line, the low power socket can be very, very small.

My understanding is that USB is only _required_ to provide 500mA * 5V worth of power and that the device must achieve some sort of functionality at this power level (possibly only enough functionality to show an error and state that it needs more power).

There is a standard protocol for the device to request more power from the USB host and the upper limit on what can be transferred is reasonably high. An intelligent charger could easily use this protocol to charge at whatever rate that the charger can handle.

By the way, I've always thought that the 50VDC standard for phone equipment would make an excellent power supply for household electronics that could be combined with an ethernet-style LAN connection in a single plug. It would be tolerant to 48V and anything around the region that a standard 48V lead-acid battery is likely to cycle over (50V is typical trickle-charge top up voltage). Probably not ideal for mobile devices, but great for desktop equipment. It's difficult to electrocute yourself with 50V DC, so that must surely be attractive. But then again it is also surprisingly hard to extinguish a DC arc, especially when driven by a backup battery and filled with ionised copper, so the phone equipment standard has at least one downside.

When I say, "extinguish a DC arc", that is not really related to the ability of the insulation to resist breakdown at 50V. High quality insulation is easy, check your local ethernet 100Base-T transformer which is rated to 1500V isolation and in a tiny package too. The problem is not so much avoiding the arc in the first place, it is what to do on the day you do get an arc (for some stupid reason, which always happens) and then you have the completely different problem of putting the arc out. Talk to someone who does welding for a living :-)

AC power was a great idea back in Tesla's day, but modern electronics shifts the advantage back to DC power (Edison finally gets it right). However, the hidden safety factor of AC power is that arcs collapse easily.

Add new comment