I was looking at a power supply on a big server box the other day and noted just how much power it seems to be wasting (presumably as heat as opposed to some other means like microwave radiation or flashing lights). It had a listing with two voltage ranges, 100V-127V at 11.6A and 200V-240V at 4.8A.
I've seen several other power supplies, mostly from online tech specs, that either show some level of increased efficiency at the higher voltage(s), or at least no difference. Or is this really efficiency? Perhaps could it be the case that it simply can provide a little more current on the DC side when operated from the higher voltage? Any possibility that it's really a frequency issue (e.g. more efficient at 50 Hz for some reason)?
I really think this is increased efficiency at higher voltage due to higher currents at the lower voltage causing more heating. The few samples I have seen don't really give me a good sense of how widespread this is, and how much is lost over all power supplies on average. Anyone else have any real figures for this?
The National Electrical Code used in the USA has a restriction 210.6(A)(2) that limits the voltage between conductors to 120V for circuits supplying cord-and-plug connected loads using 1440V or less, in dwelling units. This a restriction I would like to see removed, or at least made substantially lower, so that people in the USA can have circuits to power computers that can run at a higher voltage for more efficiency, if they choose to do so. Otherwise people wanting to do this would have to either violate the rule, or be clever and try to get around it by claiming the circuit is, or could be, for some appliance that exceeds 1440 VA (where that "appliance" may be a UPS that subsequently supplies the computer(s)).
Finding proper surge protectors might be an issue. In theory I could use German standard equipment because I think surge protection does not care (much) about the frequency (e.g. 50 Hz vs. 60 Hz). Certainly it would care about the voltage, and this would be quite different between Germany and the US when looking at voltages relative to ground (where surges would presumably be shunted), making such protection substantially less effective.
A side question: Given that grounded plugs are reversible in Germany, do surge protection devices protect both power conductors with redundant parts or are the devices designed to require that the plug only be inserted one way (I've heard some UPSes require this and won't start if the plug is in the opposite way).
There's virtually no market for 240V wiring in the USA. Special circuits have to be installed to get that voltage and it won't be commonplace to get them already in place unless many appliances use it. But the appliances won't be made because almost no one can use them due to lack of circuits. That much is a common problem in deploying something new. The NEC rule is also in the way, and that probably needs to be changed before any of this can be possible.
Under the concept of energy efficiency, many jurisdictions in the USA are now pushing toward more energy efficient lighting like fluorescent lights. Unlike incandescent, these lights actually work better on higher voltages. There is also the NEC rule 210.6(A)(1) that blocks this. That one I think should be rephrased to apply to incandescent only. Inductive ballasts do come in multi voltage (separate wires per voltage). I presume electronic ballasts can be autoranging and handle anything up to 277 volts easily.
Probably the biggest issue in enabling more use of higher voltage in the USA is the variation. Some homes have 240 volts while a few (typically apartments and condos in large buildings) only have 208 volts. But at least this can still be handled by the few types of loads being considered for this (e.g. fluorescent lights with autoranging ballasts and computers with dual voltage or autoranging power supplies). The costs of switches for the fluorescent lights would be higher due to the need to switch both current carrying conductors, but once in mass production, they should be less of a cost increase than the up front costs of changing incandescent to fluorescent.
One side benefit of this is that more common availability of US style 240 volt circuits would reduce the need for special circuits under Article 647 of the NEC (these are the balanced power circuits with 60 volts between hot and ground and 120 volts between both hots). A lot of audio equipment now comes capable of international voltages these days.