Home: Electricity: Power Outlet
Smart power outlets   (-3)  [vote for, against]
Let appliances negotiate their power requirements

Triggered by the 12 volt house idea.

New smart appliances send a request for a specific current & voltage to the wall outlet. The wall plate receives this & does a combination of Pulse Width Modulation & AC phase combining in order to produce a regulated DC current of almost any voltage.

This eliminates the need for a variety of wall warts, and allows much smarter power management decisions. Since PWM is more efficient that transformers, you'd also get some power savings right away. The big win would be that a microwave can now put itself on standby & draw lower power until it's ready to be used again.

The catches that I can see are that extension cords get complicated & you might need to run 3phase power throughout your house. On the other hand, you could the entire house on a 599 volt circuit and the correspondingly smaller internal wiring.
-- tenhand, Jun 04 2001

baking, from the appliance side http://www.wired.co...mustread.html?pg=10
wired mag article re: "energy vampires" and a solution to the problem. [mihali, Jun 04 2001, last modified Oct 05 2004]

Power socket/cord standards. http://www.powercords.co.uk/standard.htm
[StarChaser, Jun 04 2001, last modified Oct 05 2004]

Disclaimer: I am not an electrician / EE.

Doesn't pulse width modulation involve hacking off the end of a sine wave, making the power 'noisy'? Say the thing asks for 60v instead of 120v, PWM then stops the alternation of the current halfway through. Some dimmers work like this, and make all kinds of noise on radios and things.

Of course, it's also possible I'm completely wrong, and this is a different way of doing it...
-- StarChaser, Jun 04 2001


Noise should be possible to get around, since most AC appliances would want full bore current ( and no PWM) when running . The wall warts we're trying to replace are almost always DC, so the buzz could be filtered out. Power saving modes shouldn't need clean power unless they're using the AC as a timer. Because we're doing this by magic, we'll have all of the PWM devices switch on 2.6 Ghz or some other frequency that's already helplessly noisy.
-- tenhand, Jun 05 2001


There's a disadvantage to doing this. It means that the electronics inside each appliance must be a lot more complicated to "negotiate" what voltage, etc it wants. In effect, you're going to force every device to have some kind of microcontroller in it to tell the outlet what voltage it wants. That may end up being more expensive that a $20 wall-wart to power it, although it's a nifty idea. And if the automatic voltage thingy screwed up, you'd have a dead appliance.

I'm not sure whether this idea would be a good thing or a bad thing for someone experimenting with electronics.

Is this PWM that you're suggesting just switch-mode power supplies in disguise (i.e. the same type used in computers and many other beasts)? Modern switching power supplies are a lot nicer and cleaner to their surroundings. When was the last time that your computer interfered with your radio?

I didn't think that microwaves used very much power when they weren't in use anyway - just enough to power the clock and whatever nifty digital electronics people love to build into modern appliances.

Disclaimer: I know a lot less about electronics than you might think ;)
-- cp, Jun 05 2001


I meant noisy as in radio interference, not buzzing in the power...The things that require wall warts also require DC, so your system is going to need to be able to do that, too.

Wall warts are cheap, and make the electronics somewhat cheaper because you don't have to add in a specific transformer, and by a simple change of warts you can use it on different voltages. Aside from being a pain for taking up a lot of space in power strips, there's nothing wrong with them...
-- StarChaser, Jun 05 2001


Nothing wrong with wall warts except their inefficiency & the fact that they draw significant power when the appliance is turned off.

My old microwave drew between 7 & 10 watts when only the clock was running. I have a small tape player that draws 12 when turned off. In both cases they were using the same transformer for the clock & the big loads. These things add up when you consider you have a bunch of them in the house & they're running 24x7. You can probably save $5-10 over the life of the appliance fairly easily by subtracting 5 watts from the phantom load.

The appliance side negotiation chip should be about as cheap and simple as USB. I think that's about $3 now adays. And you could do a slow start to make sure you don't fry anything. If the appliance doesn't ask for more power, it won't get it. (Of cousre the first metric<-> imperial or endian problem is going to cause exploding toasters)

And no, I have no idea how a PWM circuit is built, I just know that it's one of those things that reacts well with moores law.
-- tenhand, Jun 05 2001


Would it be possible to do something similar to this mechanically, through varying the length of the prongs on the plug? For instance, something that needs only 12V has plug prongs that are 1/4 inch long, and only touch certain contacts in the outlet, but a 120V appliance has 1 inch long prongs that touch additional contacts in the outlet which provide the additional voltage. By the way, I have no idea what I'm talking about.
-- PotatoStew, Jun 05 2001


when appliances are in standby mode i.e. "off" they still drain a significant amount of power. in fact, approximately 10 percent of all electricity goes to supply standby power, according to a study done by researchers at UC Berkeley and the Lawrence Berkeley National Laboratory. see link for the full article. could this be part of the problem in california?
-- mihali, Jun 05 2001


I was going to be smart-assed and ask how the appliance could tell the wall-socket what voltage it needs when it hasn't been given any yet (chicken/egg), then I thought of having a mains-voltage circuit in the appliance which generates the required low-voltage which it sends to the wall socket. The wall socket then uses this voltage to generate a signal to modulate a PWM circuit which then supplies the required voltage to the appliance. The appliance then disables the mains-voltage circuit and runs off the low voltage supply.
Of course, you still need a switch mode power supply in the appliance to generate the 'indicator' voltage. Oh, well.
Disclaimer: I know a lot less about electronics than [cp] might think.
-- angel, Jun 05 2001


This doesn't make any sense.

There's nothing you can do in the socket that you couldn't also do in the wall wart. If "PWM" is so much more efficient than a "transformer", then do PWM in a wall wart (or inside the appliance). So, this cannot possibly lead to *any* efficiency gains -- so all this talk about "energy vampires" is purely tangential, right?

(I suspect we have another voltage/current confusion.)

So what is the benefit? Well, as with the 12V house, you replace appliance/gadget power supplies with fixed infrastructure. That means fewer wall warts bonking arround. But general-purpose, high-power, variable-voltage power supplies that can serve any need will be expensive (even in bulk), so the cost of wiring a home will go *way* up.
-- egnor, Jun 05 2001


egnor: yes the "energy vampire" talk is tangential, but i thought that the chip that is mentioned in the story might be something similar to what tenhand was implying, only on the appliance side, not the house side.
-- mihali, Jun 05 2001


I've had this idea too, mostly as a way to eliminate wall warts. (My variation on this is for the fancy circuitry to be in the equivalent of a power strip; that way it could be shared among several devices. It would also be a universal interface: all devices could use the *same* power strips / wall warts. The wall warts would be more complex, but you'd need fewer of them.)

It's not hard to build an efficient always-on supply that can only provide a few milliwatts. That's enough to power the chip in the appliance, which can start up the negotiation for real power. Power draws could be renegotiated, as tenhand says, when the appliance changes mode (eg from the not-quite-"off" mode to fully-"on").

The appliance-side circuitry to do this needn't be expensive --- well under a dollar should be achievable using current technology. Look at how expensive a low-end PIC is these days: 94 cents. Less, I'm sure, if you buy 10,000 of them. The physical power connectors probably cost almost that much. If this were widely adopted it wouldn't be very expensive.

I envision two interfaces --- one for lower power devices (less than, say, 25 watts), and one for higher power devices (which would need bulkier connectors). Actually, I'm not sure if the higher-power interface would be useful. You might as well just build the phantom-power stuff into the appliance at that point.
-- wiml, Jun 06 2001


I'm with [PotatoStew] on this one. One could make key-like prongs and supply voltage based on the shape of the prong. This eliminates the necessity for any extra intelligence in the appliance itself. Of course, as everyone else is saying, the real issues lie elsewhere.
-- phoenix, Jun 22 2001


That being baked, already. || prongs get you 120v, |- gets you something else, and three curved prongs in a triangle gets 220v. <That's all I can remember, but the link above has a little more information.>
-- StarChaser, Jun 23 2001


StarChaser: You're probably thinking of two-phase AC.

Anyway, this is a hugely inefficient idea.

First: If you decrease the voltage at a central place within the house, than, to transmit the same amount of power, a larger current will need to flow through those wires. That creates more resistance losses, your wires get hotter, and your house burns down. It is much more efficient to transmit power at high voltage and low current than the other way around. So wall worts are actually MORE efficient.

The ONLY problem with wall worts is that many are poorly-designed and essentially shot-circuit once you open the circuit attached to the secondary coils. The solution to this isn't to do away with wall worts, but to stop producing bad ones.
-- TerranFury, Feb 13 2004


Wall warts don't "essentially short circuit" when the output is open; a wall wart will draw less current when the output is open than when it's loaded. The amount of power wasted is often greater when the output is open than when it's loaded, but I believe that the amount of power drawn from the wall increases monotonically with load for most cheap wallbrick designs.

That having been said, if a wallbrick draws 6.5 watts from the wall to feed a 6 watt load, versus 4 watts when it's feeding nothing, the difference in loaded versus unloaded proficiency can be very profound.
-- supercat, Feb 13 2004



random, halfbakery