Rethinking household/office power, beyond 60hz


I've written before about the desire for a new universal dc power standard. Now I want to rethink our systems of household and office power.

These systems range from 100v to 240v, typically at 50 or 60hz. But very little that we plug in these days inherently wants that sort of power. Most of them quickly convert it to something else. DC devices use linear and switched mode power supplies to generate lower voltage DC. Flourescent lights convert to high voltage AC. Incandescent bulbs and heating elements use the voltage directly, but can be designed for any voltage and care little about the frequency. There are a dwindling number of direct 60hz AC motors in use in the home. In the old days clocks counted the cycles but that's very rare now.

On top of that, most of what we plug in uses only modest power. The most commonly plugged in things in my house are small power supplies using a few watts. Most consumer electronics are using in the 50-200w range. A few items, such as power tools, major appliances, cooking appliances, heatters, vacuum cleaners and hairdryers use the full 1000 to 1800 watts a plug can provide.

So with this in mind, how might we redesign household and office power...

The first assumption I would make is that we would like smart power. With smart power, a data protocol would exist between the power supply equipment and the power using equipment (or "load") and/or the recepticle in the wall. The supply would talk to the load to learn what sort of power it wants or can use, and only provide full power when the load is properly connected. (There would always be some basic very low level power available to run the devices that do this data protocol.)

In the quantities of which we are speaking, these devices would be very cheap -- pennies -- the way USB chips have become. Default chips for legacy devices (which would just say, "I need 120v at 60 hz 15 amps" or similar) would cost just a few pennies, and adapters to plug old devices into new, smart recepticles, would be made very cheaply." Adapters to plug "new" devices into old sockets would also be cheap, though perhaps just a couple of dollars.

The base power transmitted would possibly be high-frequency AC. That's because most devices these days, with switched mode power supplies, immediately convert the incoming power to high-frequency AC. (This may not be able to make universal. Many power supply designs regulate the output voltage by controlling the frequency of the intermediate AC.)

The main supply, down in the basement, might know things about the wiring (be it old or new) including gauge, resistence, length and voltage rating. Some of those things it could measure with the aid of a smart endpoint.

There are trade-offs over what voltage you use. Low voltages are safer, and of course are what most modern electronics actually use. Low voltages require high current (and thus thicker wire) to deliver any real power. High voltages can deliver lots of power over thinner wires, but can also do damage to people and property. They need more insulation and larger spark gaps.

Smart power allows higher voltages for several reasons. First, we can have ground fault circuit interrupters, which are already not that expensive, to assure that power is shut off if it ever leaks out or could cause other trouble. With smart recepticles, it is possible to determine how much power is being lost in the internal lines to spot arcing or overheating. In addition, with smart power, high voltage is never applied until the load is properly connected. Your kid can stick a knife into the wall socket at no risk.

This might allow the wires to the oven, clothers dryer or dishwasher to run at 600 volts or even more depending on the rating of the wire. That means several times more power over the same guage of wire. Or lower current for the same power with less loss and risk of melting wires. The special lines to your electric car might like even more voltage and would be wired for it.

DC devices would often not have power supplies, or just have minimal rectifiers and a low-drop voltage regulator. They would just plug into the wall and ask for the voltage or voltages they need. Or at the worst, plug into a universal supply that plugs into the wall.

With low voltage it might also be possible to construct walls with power "strips" that allow the placement of recepticles anywhere on the wall. To do this, two horizontal metal strips would be laid like a "bus" along the entire wall at socket height. These would, on demand, provide something like 12v at 100hkz AC (just an example frequency, the right universal frequency would be carefully chosen for best utility and minimal interference.) Later, after the wall is constructed, one could cut a hole just above the strips for a smart power box, and punch lines to the strips, then plaster and paint over.

It's also possible to imagine a power wall. In this case, there would be two planes (live and ground) sandwiched over an insulating layer. All low voltage, and of course protected against shorts or ground faults. Possibly with extra layers to reduce interference. Then one could place devices (clocks, flat panel screens, efficient lights) anywhere on the wall and punch through and clamp to get power for them -- no wires. Of course such a wall would block radio signals, which is a downside. You would still need to wire some sockets for high power devices like vacuum cleaners, independent of this power wall.

Power wall material might also make sense to build into desks and equipment cabinets. The material could come with pre-drilled holes that can be found with a sensor and punched into by anybody, or one could drill a hole (with power off) and then insert a standard plug which clamps onto the two planes to securely connect to the power. Of course all devices on the power wall must use the same base power, so some will need converting power supplies.

It would also be possible install these strips or planes in floors and ceilings. If the metal for a plane is too expensive, strips (which again can easily be located with a metal detector or AC sensor) could be laid on studs. You could not put devices literally anywhere, but you could come close.

And of course, you would never have to carry power supplies for your portable devices once this caught on. There would be power for them everywhere. Before the wall sockets get smart, smart universal power supplies would be bought which plug into regular household voltage and deliver the power needed, as I described in the earlier post.

How to get there

Here's the path I see to take us to a smarter power world.

  1. People start creating more universal DC power supplies, the type you can already buy with smart-tips that tell the supply what voltage to provide. A protocol is defined for the tips to be active.
  2. These supplies get cheaper, and we start seeing "multiple" units that will power many devices at once with different tips, just to replace the nest of power supplies we have now.
  3. People start making devices with a smart power plug on them. Universal power supplies are provided or sold as an accessory with them.
  4. Eventually, devices stop coming with a power supply. You're expected to have one of your own.
  5. A universal switching supply for DC devices is available to be installed at the power distribution center. People doing new home construction start using it. Others use it for power-over-ethernet based power. This gets cheaper.
  6. Larger units capable of providing high power become available, and are installed in new construction. Cheap adapters are put in place to have wall sockets generate 110v as desired for old devices.
  7. More and more devices become able to connect to the smart jacks, no need for adapters.


Q: How did God create the world in seven days?

A: He didn't have to worry about an installed base.

Obviously you need to do this but their needs are simple. Clearly one of the primary things you have is an ability to plug in regular household voltage devices, and tiny cheap adapters that put out the DC voltages needed by our many DC devices.

I have a universal laptop adapter, for example. It takes many tips, and puts out different power based on the tip. I haven't examined it in detail, but the tips have 4 wires, so I presume the other 2 wires have something as simple as a resistor which the PS reads to figure out what tip is present and what voltage to provide. This can obviously be done quite cheaply. However, if you look at the price of USB driver chips today, you see that an active device can be very, very cheap too.

DC power has many problems as a household power distribution system, and there may be other answers to the wart problem. You need to consider the effects of wiring, efficiency, etc. There are good reasons why even remote PV powered residences use AC distribution within the home.

I was asked a similar question at a meeting about LED lighting. LED lights use 5V DC and are roughly the same efficiency as flourescents. So a normal bright room light needs 20W (it contains multiple LEDs). At 5V that is 4 amps.

Power loss due to resistance goes as I^2*R. For a 100ft run of 12 gauge copper (typical 20A rated home wiring) R is 0.3 ohms. If you operate at 5V that one light would consume an additional 5W (16*0.3) to heat the wires. If you move up to 10Amps (50W) you also consume a whopping 30W heating the wires. So DC wastes 20-50% of your power. That same 20W light on 200v AC has a wire loss of 0.003W, which is quite insignificant.

For more power hungry devices like PCs, microwaves, or air conditioners a 5V system would need 250amp service. That is utterly impractical. So you will also need 100V+ service in the house. Dual wiring is expensive so you need really compelling reasons to add DC.

AC has other advantages. It is easy to shift AC voltages with transformers. DC to DC voltage shifting is very hard for up, not so hard for down. Internally, CRTs and microwaves need 10-50KVolts. Other common needs are 40v, 10v, 5v, and 3v. It's easy to get all of these with transformers.

Similarly, variable frequency AC is superior for motors. DC is very simple, but unreliable and less efficient. Improvements in power semiconductors have made variable frequency AC motors affordable. They are much more reliable and efficient than DC motors. Railroad locomotives have largely shifted over to variable frequency AC and the proposal for next generation automobiles is variable frequency AC.

The automotive proposal is for the frequency to be up in the 5-20KHz range, and the voltage to be approximately 40V. That lets them replace hydraulics (brakes, steering) and mechanicals (air conditioner, valves) with electronic components. The variable frequency AC components are impractical without modern power semiconductors, but they are also more reliable and more efficient than the hydraulic and mechanical components that they replace. For the devices that need DC (radio, LED lights) the higher frequency AC lets the AC to DC converter be much simpler. It can be put onto a single hybrid chip and integrated into the LED socket or electronics card.

For the home market you have the problem that higher frequency AC has many problems for power distribution. The 50-60Hz AC is really very good for the metropolitan power distribution efficiency. The low frequency minimizes resonance losses to the earth.

The change that I see is different. Some of my newest equipment has AC to DC converters that are no larger than a typical 3-prong plug. They generate variable voltage DC (6-10v). It looks like a poorly filtered power supply, and is probably a diode bridge with modest filtering. That keeps the analog components small. Then there is a DC to DC down converter to 5V or 3V within the device. This kind of splits the power supply into an small AC cord part and a small device part. For power levels up to about 10W this size reduction seems feasible. Above that you need some breakthroughs in analog component sizes. If you pick just a few amperage ratings, e.g., 0.1A, 0.5A, 1.0A controlled off plug shape you end up with a simple DC powering environment where three common cords support any device that needs up to 10V and up to 10W. The device half deals with the particular device requirements.

The USB path is a dead end for higher power requirements. The tiny wires in the USB cable have a high resistance. They limit the USB power to under 0.5A for a reason. If you push it much further you end up with hot wires, significant voltage loss, or both.

I am not proposing DC for high power devices, though I think it is suitable for low power devices. Most of what I have plugged in in my home are low power devices with power supplies. Higher voltages are required for higher power, and even higher than 120v makes sense was part of my point, with modern tech.

I am not proposing high-frequency for long haul though it might make sense for short-haul (from transformer to house).

For many devices, high-frequency modest voltage would be easy to convert to the power needed. At a minimum you would not need to do the rectify and chop portion of the typical switched mode supply.

Many of the things that make sense in the car can make sense in the home or office.

It would be nice if plug in low power devices would all centralize on a voltage, but we can't seem to get them to do that. But maybe it can happen.

The goals that I see are

  1. Be highly energy efficient
  2. Replace the many custom power warts with one generic power cord

I think that we will be stuck with 100-240V AC distribution within the neighborhood and within the home. This meets the needs of the many high power devices and is the installed base. I had relatives who participated in the conversion of Boston from DC to AC. Given the immense work that it took then I cannot imagine replacing the current AC distribution systems. The need would have to be earth shattering.

For low power devices (under 10W and under 10V) I do expect the outcome to be a split power supply. The generic DC cord will provide up to 1A of variable DC. Then each device will take the variable DC and internally provide regulated power at the needed voltages.

This is much more energy efficient than any low voltage distribution will be. The generic DC cord can be made efficient, although the combination of generic DC with custom regulation is probably not a match for the best switching power supplies. But much of the waste in the low power devices is leakage waste, and that could be made quite low.

I have three devices that are close to this:

  1. My antique Psion 5, whose custom power adapter is about the size of a 3prong plug.
  2. My Zaurus CL3000, which has a really tiny little power adapter. It's 4x4x1.5 cm. It provides 1A at 5V nicely regulated. The size is a bit misleading because the prongs are pivot mounted and can fold into the supply. When folded, the whole thing is just 4x4x1.5
  3. A mystery adapter. I wish it were properly labeled. 1x3.5x1.5 inches, producing variable DC (5-10V) and accepting 45-65Hz 100-240V.

A standard 3-prong plug is 2x1x1 inches. All of these exclude the actual prongs. So current products are close to meeting the size goal. I expect that the generic 10W DC cord will emerge. It can be introduced fairly easily because on one end it can use the existing 45-65Hz 100-240V power grid, and the other can be a generic 10W DC.

You could push the power rating up further but when you leave the domain of pure electronic devices and introduce motors you gain some difficult power supply issues. There are many reasons for the custom voltages when driving motors, and usually the cost tradeoff between custom voltages and difficult mechanical design ends up with custom voltages.

We are indeed stuck with the electrical specifications our power companies send us. Here in the U.K. (and most of Europe) we are provided with 230vac @ 50Hz single or per each of three phase, mostly protective multiple earth (P.M.E.) which means the neutral wire is strapped to the earth conductor at, but not after the incoming main cable terminus. Transforming these voltages is well understood and inexpensive.

In an immoveable house the need is for high-speed very low cost data /voice transmission wherever it is required. WiFi even with MIMO is fine (or will be) but, to this writers mind, it is much better suited to a mobile environment, and it in no way eases the problem of delivering energy to a device. We surely don’t want to run everything on batteries.

Rethinking house wiring probably requires the delivery of power and data simultaneously on the three-wire system we are already used to.

Two types of power are required, medium voltage (up to 400vac) and low voltage (up to +/- 50vdc). The international standard telephonic voltage is minus 48vdc (chosen when Henry Ford was friendly with Thomas Edison!). The rationale for this choice was: - 48 is a multiple of 1.5 (that voltage produced by a single Leclanché cell, and minus to earth potential so as to prolong the life of exposed copper wire on poles by causing electron flow to travel to the copper (clever eh?).

One type data transmission is required, the collection of packet switched protocols known as TCP/IP at speeds fast enough to cope with digital TV and VoIP. A Protocol now exists for data superimposed on power cables, so there is no need to reinvent the wheel.

The benefits of a combined system are legion: - no more Cat.5, TV or phone cables all over the house, power (medium or low) at every outlet, only one lead to any device, one can still have WiFi if one wants, much lower total cost of installation, the use of co-axial cable instead of twin and earth, the use of a ring-main (almost unique to the UK) could be continued, the Live conductor need not be connected if not required, etc.

One can envisage a mains consumer unit with connections for TV antenna, telephone /ADSL line and WiFi output to ones mobile devices (cell-phone, car, boat, bicycle, pram, etc.).

To invite further discussion, what would be the problems with providing say 230vac between the live and neutral wires and minus 48vdc between the neutral and earth wires? How would these voltages interact? Could any interaction be easily overcome? Only 2 wires in the USA?, one at plus 110vac the other at minus 110vac?

Ian Robertson

I appreciate all your article, they are all very practical and quite useful for many of us. Rethinking the electrical system might be one of those ideas that can generate a new start. As the matter of fact this should be the main line for our protection and for the environmental protection. I don't see this happening any time soon but the least the we can do is to act smart when it comes to energy consumption. Household appliances have a major importance here, people need to understand how thing go. I just bought myself brand new washer parts and I can tell the difference on my electric bill. But it's not all about energy consumption it's also about radiations and many other factors that can harm us. Now I am thinking to replace all my appliances with "smarter" ones.

Add new comment