The part I don't understand is that the majority of the devices in our houses use DC. [The exception would be things with AC motors (fridge and dish washer) or heaters (toaster ovens and clothes dryer).] The worse part is that most of these convert AC->DC quite inefficiently. So I know I wouldn't argue "costing several times more" without it worked out. Also figure in that regular incandescent lights are horribly inefficient compared to white LED lights (DC, of course). DC can drive motors (unless you're talking 3 or 6 phase motors) and make heat just as well as AC. Perhaps transmitted AC is better, but it appears to be the wrong choice for the common household...
Well, don't blame Tesla for incandescents. >So I know I wouldn't argue "costing several times more" without it worked out. (I^2)*R wipes you out in a DC transmission line. Forget about transmission losses, you couldn't afford the copper. Without transformers the grid would be a worldwide space heater.
I wouldn't transmit DC on the grid, but maybe have it available in-house on a central transformer... Common AC->DC transformers still use some power when you're not even using it. They're starting to make electronic transformers that are more efficient and off when not in use, but they cost more so they're not widely spread... DC Loss is still worse than that for most purposes, you'd need higher gauge wire which will cost more, and you need to regulate the voltage at your device. Your 120AC will be about that at the outlet, but 100DC (or whatever) is gonna vary throughout a house
Oh, I'm not blaming anyone. I'm not saying DC should be used for transmission (but I vaguely recall HVDC lines are best for *really* long distances) lines, but that it doesn't appear to be the best choice for households. I suppose DC is even better placed in server rooms and server farms. Why taken 20 hits at 50% efficiency, when you can take one hit at 90% and some resistive losses? It's been a while since I've delved into the nitty gritty of, but whatabout the Real{VI*} losses from AC? You still have resistive losses...
Well, you don't need a transformer for AC->DC conversion (full-wave rectifier and a good buck DC-DC converter). I know of several really efficient power supplies (most notably Xantrex) that actually do AC->DC->AC->DC for optimum noise/ripple reduction (DC->AC stage is at several hundred KHz and not a harmonic of the original AC). "Real power flows on phase change and reactive power flows on voltage drop" right? So the reason you don't get much of a voltage drop with AC is because most power usage is resistive. Going into a hypothetical place where a house is given 240 VAC and distributes several DC voltages (let's say 240 VDC, 120 VDC and 24 VDC). Low power devices would take 24 VDC (computers, alarm clocks, dvd player) and high power devices would take 120 VDC (amplifiers, toaster oven) and really high power devices would take 240 VDC (oven, clothes dryer). The high and really high power devices don't need much in terms of regulation. Most low power devices would most likely need to do their own dc-dc reduction anyway so regulation is a side effect. DC-DC conversion can be really efficient. I know of one design that is 99.5% efficient at 400 watts and drops down to a mere 95% at 1100 watts. Then again, it cost several hundred dollars but...... (You guys do realize that I'm justing thinking out loud and am not an extremist who thinks we should all have switched to DC yesterday, right?)
What about the fact that AC is at regular intervals so running an accurate clock would be very difficult?
I'm confused.... Why do you need AC to sync a clock? What kind of clock? Like a time clock or a CPU clock? Either way, you don't need the periodic nature of AC to have an accurate clock. To prove my point, look at a wrist-watch. Those things, for as cheap as they are, have absolutely phenonimal accuracy (seconds over a period of a year) and don't rely on AC. Crystals and oscillators are much better for accuracy.
Actually, AC is better, the electrical companies know that most households use clocks timed off AC, and if they get a bit ahead with their timing, they slow down the output until it is back to accurate.
As a EE, you two are thoroughly confusing the heck out of me!!!!!!!!!! Might as well be speaking gibberish, cuz it wouldn't make a bit of difference.... Where are all these magical clocks? If you ask me, clocks are a pretty pathetic way to argue that AC is better....
The signal that is sent through powerlines is at a certain frequency (60Hz). Clocks that you plug into a wall step down that frequency to 1Hz, and use that to time a second accurately. This is cheaper in a whole than using crystals, and also alot easier to utilize.
I would disagree about being cheaper and easier, but then again I don't make clocks. Since the basis of using them is accuracy, I don't buy that either. Using a 1 MHz crystal, you get 30 ppm which puts the frequency at 1 MHz +- 30 Hz. I guarantee this is better (30 ppm) than what the power company can give you. The only way I could see this being cheaper is if you use a small AC motor and put a 1:3600 gear reduction on it and powered it straight off the line. I personally would pay the extra $2.50 for a better clock.... Or better yet, build my own clock.
I am not sure how expensive quartz crystals are, but to step down the Hz you just need two decade counters, one set for 6 and one set for 10. Very very cheap. The decade counters can also be plugged dirrectly into the 7 segment decoder to display the time or in older clocks the signal can be brought down another 60 times and the signal flips the numbers. All in all very cheap and easy to use
Crystals are under a dollar (depending on quantity, distributor, speed, etc.). If you bought a crystal and a really large counter (>64 bits) then all you need, in addition, is the display. Better yet, just buy a chip that's intended purpose is real-time clocking. Man, talk about being off topic.
Speaking of being off topic, I was wondering if anyone knew where I could get a Quartz IC for a battery powered clock, and also how to use it. I posted this question in the Electronics forum under the sticky "Linears IC Workshop", but as this thread has gone in this dirrection, I might as well post here too.
The mains frequency does vary, but (speaking for the UK but I imagine the States is the same) the company does deliberately speed up or slow down to compensate in the long term. Certainly my various mains-operated digital clocks (cooker, microwave, central-heating, radio-alarm, etc) are always near enough to GMT, the battery-operated crystal-controlled ones drift quite a bit over a month or less. The TV and VCR clocks keep perfect time, get their time signal from a broadcast standard - way to go!
not to get back on topic, but this is fresh of /. so it may be /.'ed for a while http://www.edm.net/~jwilliams/tesla.html a cool snippet
random question: my science teacher had something he called a tesla coil and it looked like a soldering iron. he would plug it in, hold it by the metal tip, and turn it up really high, then zap us with his other hand. I know this is how tesla coils work, but there was no bulbous piece visible, just like a fat little soldering iron... I have wanted one since, and have never been able to find them. (they'd make for some great practical jokes!)
