All over the place? The only references I've seen are for a few very-long distance applications. I could be wrong, but it seems that apart from those special circumstances, the cost of the equipment to step it up and down is prohibitive, and good old AC is used.
Because when you need a conductor with a cross-section of a square inch, it's a lot cheaper to pour copper into a mold and make a bar than to draw out and assemble all of the individual strands - and at that thickness, even with very thin strands (which make it even more expensive), your cable doesn't have any significant ability to flex or bend over short distances, which is the primary advantage of a cable over a bar.
Besides, it's also easier to tighten a bolt through the bar than to have to tighten a clamp onto the cable, then the clamp to the device.
As I recall, Edison envisioned small, local power plants in each community, negating the necessity for long-distance transmission and associated losses. Provided that the power plant is not polluting in nature, that's still not a bad idea - although at least *some* provision for long-distance transmission (in case of power plant outtage) is still needed.
You can't just say that DC is better for distribution, because stepping power up and down is an integral part of distribution, and that's just not so easy with DC. If you don't believe me, our building pulls (on a normal day) close to 200 amps on each of 3 phases of 440-volt power. Why don't you design me a DC regulator that can step 440 volts down to 48 volts, passing nearly 300 kilowatts along the way?
Oh, and while you're at it, make sure that all of the circuitry needed to do that is sufficiently bullet-proof to last at least a few decades, with nothing more than an *occasional* check of the oil.
Once you've designed said device, total up the cost to build it. Get back to me.
"Big 5v/12v power busses wouldn't even need to be insulated."
Wow. Wait until someone is tightening a screw on one of them, and his long-handled ratched hits the other 12V line, with a few hundred amps coming down the pike. Unless they're *really* far apart, you're a lot better of if they're insulated. It doesn't have to be a ratchet, sooner or later, something (even something bizarre) will happen, and as long as you're not the person holding one of the pieces of metal (and your equipment isn't at stake), it's quite fun to watch.
"DC is way better than AC for very long transmission lines. For lines longer than 1000 miles, AC results in very unfavorable reactances and even radiation resistance. So, if you want to build a power line across the top of Siberia, you should use DC."
Alrighty, then. You want to build a power line 1,000 miles across Siberia. Since you don't want to pay for enough aluminum or copper to run a wire that's three feet in diameter, you have to step the voltage up very, very high in order to get it that long of a distance. So, you've got a 20KV line. If it's DC, just what do you propose you use at the far end to step it back down to something useful? Yes, it can be done, but when you look at what it takes to get it done, it's not quite as cut-and-dry.
"There are some advantages to operating on 48VDC, but unless you have the big battery room of a telephone central office, they're not that big."
You already do, in the UPS systems. In fact, you've already got an AC->DC conversion system in the UPS (although not of a large enough size, but that's the easy part), the batteries, and then a DC->AC inverter large enough to power all of your equipment. That means that you not only have the batteries, but even MORE circuitry (for the inversion) than you would with DC. More circuitry, more losses.
"Somebody, preferably a licensed electrician, has to check all the data plates, add up the current loads, calculate voltage drops, size the wire and breakers, and torque the big screw terminals to the correct torque, using the correct lockwashers. Every time you add or change a load, somebody has to recheck the math. Errors can cause a fire."
And how is that different than A/C? Look in the power room of a data center. You find distribution boxes filled with (gasp!) screw terminals, which must be torqued - and re-torqued periodically. Someone has to keep them with code, add up currents, deal with safety factors in cabling (especially over distances), and everything else. What's that? A/C systems have breakers to protect you? Well, guess what, DC systems can, too. It really isn't much different.
A/C has other problems as well. In fact, our building management has jumping through hoops trying to figure out why every CRT in our building started "shivering" around noon three days ago. They've been toquing down neutrals, hots, and grounds throughout the building, measuring loads on seperate circuits (and phases), re-distributing loads, walking around with gauss meters, unplugging various inductive devices, scouring the buiding for new installations, all of the fun stuff. If I had an oscilloscope, I'd take a look, but I don't, and that's they're job anyway.:-) The LCDs don't have a problem, as it's massively easier to filter and condition the DC coming from the wall-warts, and of course, they have the huge benefit of not having to closely regulate changing magnetic fields to keep electrons where they should be.
(None of that is to say that DC is necessarily better, just that a few of your arguments don't hold water.)
It's not the rectification, it's the regulation and filtering. Just because you've ran your 120 VAC through a four-diode bridge doesn't mean that you're ready to pipe the result into your CPU. Not only do you still have a very high voltage, you've still got the sinusoidal waves, they're just all on the same side of ground. You have to regulate the voltage, in this case done (for efficiency) by switching on and off really quickly to obtain an *average* of the target voltage. Once that is done, you have to smooth that out so that you really do have a fairly clean, stable voltage.
That's why when you open a power supply, you see switching FETs (with a heat sink), capacitors, and inductors - with the associated circuitry to make it all work, mostly switching the FETs on and off correctly.
The question is whether 48V DC (which is the most common voltage for this sort of thing) is enough to make your muscles tighten. Under normal situations, it's not.
At this point, some idiot will pipe in with "but it only takes xxx milliamps to activate your muscles, and xxx milliamps to kill you!!!!" Yeah. Volts and amps are not the same, and they're interrelated based on the resistance in question - and unless you're talking about certain moist parts of your body, your body has a pretty fair amount of resistance. As an example from finger-tip to finger-tip, 300 kilohms is the normally quoted standard, although I've measured higher and lower - but the amount varies greatly on the amount of callous, sweat, and other things.
In the end, you can grab a couple of 48V wires in your hand and unless you've got some really moist hands which you've sanded down recently, you won't even be able to feel it. Put them in your mouth, and you'll get a very painful kick, however.
In fact, much to the disbelief of many, even 120VAC, from one hand to the other, barely gives you a tickling/buzzing if you have an average amount of callous (for a man, anyway), and your hands aren't moist. Let them moisten up, though, and things get interesting quickly.
Simple. Because fewer of them are produced, so it's considered a specialty. Just because they're DC doesn't mean that they don't have regulation circuitry on them (at least you had better hope not!). The regulation in an A/C power supply is produced by the million, so the economies of scale kick in. The regulation in your D/C unit doesn't benefit from that sort of quantity.
There's a bit more truth to that than some people realize. One of my acquaintences was a door-gunner in a chopper in Vietnam. He showed me a picture of the side of his chopper, it had the various jeeps, trucks, and other "kills" painted on the side - and a bunch of elephants.
Because the VC sometimes used elephants to carry rockets, if they passed elephants, they were often ordered to fire on them without question of what they were actually used for. A.30-caliber bullet wouldn't normally be what you'd think of as an "elephant gun", but when you're spraying them out at the rate of an M60, you can certainly take an elephant down - and the ones you don't have hundreds of bullet wounds to deal with (or die from) later. Whenever man is involved, animals always get the shaft.
The cleaning crew in my office had a knack for knocking over my plants, so I bought a cheap USB webcam on ebay for $15 and a USB extension cable, and used some double-sided tape to stick it to the wall. I installed MVC on the machine for motion detection. The problem was solved, no more broken plants.
Well, because the task I desire (avoiding broken plants) is accomplished, I rarely even look at the images - sometimes not for months. But there have been two thefts in the company, and both times, the thief happened to walk down the walkway where my camera is, so I got pictures of them.
Where am I going with this? Just yesterday, I was speaking with the building management, and they brought up the camera - they saw the pictures from one of the thefts, and said that the pictures looked better than most of the very-expensive security systems of other clients in their buildings.
Linux's V4L subsystem makes this sort of thing very easy - all you need is a camera with a V4L-compliant driver, and you're set - there are many packages out there which will take care of the rest. I mentioned that I'm using MVC, which is an old, low-feature application. At some point, when I'm bored, I'm going to switch over to Motion, which is much more full-featured.
Kidneys are a lot different from bones. For kidneys, you need to find a very close match ahead of time, it's not the sort of thing that you can do in the limitted time span in which you can keep the kidney viable. Bone, on the other hand, is just crushed up and sent off to the donor bank, where it can be kept for much longer.
Organ donation is sufficiently in-demand that everything gets used. If you're a full donor, a squadron of doctors come in, and you're more or less left with your skin, muscles, and a skeleton made from PVC sprinkler pipe. Bones, eyes, organs, they take all the good stuff.
Dumb terminals? Given that you can stick more and more powerful chips in smaller and smaller devices - with less electrical consumption, it would seem the opposite. Right now, a very good many computing devices are so small that the largest impediment is the *input* - trying to type into the things.
"If I think I've developed a new treatment or prevention of cholera or typhoid, I'm not testing it in Anchorage, AK. I'm going to Manila, Buenos Aires, Rio de Janiero, Cairo, etc. to see if it actually works, once I'm pretty sure that the drug won't kill those it is put into outright. I can't just sneak into a water treatment center and surreptitiously route sewage discharge into the drinking water supply bypassing the filters and Cl2 treatments, etc."
If it were really that important, and it were really isolated to Latin America, then why are they testing it in Norway as well? As cynical as my argument sounds, I've worked for large companies, and I have seen first-hand how these decisions get made. The times when the interest of people at large - even poor people - is a deciding factor in the decision making process are so few that you can't believe it when it happens.
"Hmm... test the drug in the US. 1% chance of a multi-billion dollar lawsuit. The feds keep track of our results. OK, test in Nicaragua, 0% chance of any lawsuit, nobody knows if anything bad happens. What should we do?"
"We have no problem testing new cancer treatments that are on the verge of lethal for "normal" people on terminal cancer patients...why is this any different, then?"
Well, you tell me: Is there a difference between a potentially lethal drug being used on a terminal patient, and a potentially lethal drug (yes, the old one killed some people) being used on someone that's just going to have diarrhea for a few days?
Sure, you could argue that those people would die from dehydration resulting from diarrhea, but having lived in those areas, most of the people which are likely available for the tests also have easy access to pedialyte to take care of that.
Wait until you've seen Latin American cholera wings in hospitals, where the patients are in hammocks with holes cut out for their butts, and buckets underneat them. =)
I once dated a girl who worked in a pathology lab identifying contagious disease samples. When she'd talk about it, she made it sound like she dealt with some pretty serious stuff, so I asked her what she had worked with that day. I don't recall all of the different things, but among them were two different types of ebola. That kind of surprised me. I mean, the 3 or 4 different types of HIV you can expect - with hundreds of millions of people infected all over the world, that's a given, but ebola is vastly less common, and I didn't expect to hear of her dealing with two different types in a single day.
It'll get tested on them for efficacy. Once it's proven effective, then they'll get charged for it. That may sound cynical, but look at what' happened with the recent rotavirus vaccine - since the last rotavirus vaccine was taken off of the market for safety concerns, the new one was tested almost entirely in Latin American countries. The official explanation was that it was needed there more than other places - but the real reasoning is that those locations are easy and cheap to test in, with little or no restrictions, and little (if any) accountability.
It's all a matter of priorities. I'll bet that of those people not willing to pay $25 or $30 for entry-level broadband, a good portion of them spend $50, $60, or more on cable or satellite TV.
Wake me up when^H^H^H^H *if* this ever comes to pass. If I had a gigabyte for every claim I've heard that the next mammoth-sized storage technology (usually optical-based) was just around the corner, I'd already *have* my 1.2 petabytess
Slashdot Mirror
Points taken!
steve
All over the place? The only references I've seen are for a few very-long distance applications. I could be wrong, but it seems that apart from those special circumstances, the cost of the equipment to step it up and down is prohibitive, and good old AC is used.
Because when you need a conductor with a cross-section of a square inch, it's a lot cheaper to pour copper into a mold and make a bar than to draw out and assemble all of the individual strands - and at that thickness, even with very thin strands (which make it even more expensive), your cable doesn't have any significant ability to flex or bend over short distances, which is the primary advantage of a cable over a bar.
Besides, it's also easier to tighten a bolt through the bar than to have to tighten a clamp onto the cable, then the clamp to the device.
As I recall, Edison envisioned small, local power plants in each community, negating the necessity for long-distance transmission and associated losses. Provided that the power plant is not polluting in nature, that's still not a bad idea - although at least *some* provision for long-distance transmission (in case of power plant outtage) is still needed.
steve
You can't just say that DC is better for distribution, because stepping power up and down is an integral part of distribution, and that's just not so easy with DC. If you don't believe me, our building pulls (on a normal day) close to 200 amps on each of 3 phases of 440-volt power. Why don't you design me a DC regulator that can step 440 volts down to 48 volts, passing nearly 300 kilowatts along the way?
Oh, and while you're at it, make sure that all of the circuitry needed to do that is sufficiently bullet-proof to last at least a few decades, with nothing more than an *occasional* check of the oil.
Once you've designed said device, total up the cost to build it. Get back to me.
steve
"Big 5v/12v power busses wouldn't even need to be insulated."
Wow. Wait until someone is tightening a screw on one of them, and his long-handled ratched hits the other 12V line, with a few hundred amps coming down the pike. Unless they're *really* far apart, you're a lot better of if they're insulated. It doesn't have to be a ratchet, sooner or later, something (even something bizarre) will happen, and as long as you're not the person holding one of the pieces of metal (and your equipment isn't at stake), it's quite fun to watch.
steve
"DC is way better than AC for very long transmission lines. For lines longer than 1000 miles, AC results in very unfavorable reactances and even radiation resistance. So, if you want to build a power line across the top of Siberia, you should use DC."
Alrighty, then. You want to build a power line 1,000 miles across Siberia. Since you don't want to pay for enough aluminum or copper to run a wire that's three feet in diameter, you have to step the voltage up very, very high in order to get it that long of a distance. So, you've got a 20KV line. If it's DC, just what do you propose you use at the far end to step it back down to something useful? Yes, it can be done, but when you look at what it takes to get it done, it's not quite as cut-and-dry.
"There are some advantages to operating on 48VDC, but unless you have the big battery room of a telephone central office, they're not that big."
:-) The LCDs don't have a problem, as it's massively easier to filter and condition the DC coming from the wall-warts, and of course, they have the huge benefit of not having to closely regulate changing magnetic fields to keep electrons where they should be.
You already do, in the UPS systems. In fact, you've already got an AC->DC conversion system in the UPS (although not of a large enough size, but that's the easy part), the batteries, and then a DC->AC inverter large enough to power all of your equipment. That means that you not only have the batteries, but even MORE circuitry (for the inversion) than you would with DC. More circuitry, more losses.
"Somebody, preferably a licensed electrician, has to check all the data plates, add up the current loads, calculate voltage drops, size the wire and breakers, and torque the big screw terminals to the correct torque, using the correct lockwashers. Every time you add or change a load, somebody has to recheck the math. Errors can cause a fire."
And how is that different than A/C? Look in the power room of a data center. You find distribution boxes filled with (gasp!) screw terminals, which must be torqued - and re-torqued periodically. Someone has to keep them with code, add up currents, deal with safety factors in cabling (especially over distances), and everything else. What's that? A/C systems have breakers to protect you? Well, guess what, DC systems can, too. It really isn't much different.
A/C has other problems as well. In fact, our building management has jumping through hoops trying to figure out why every CRT in our building started "shivering" around noon three days ago. They've been toquing down neutrals, hots, and grounds throughout the building, measuring loads on seperate circuits (and phases), re-distributing loads, walking around with gauss meters, unplugging various inductive devices, scouring the buiding for new installations, all of the fun stuff. If I had an oscilloscope, I'd take a look, but I don't, and that's they're job anyway.
(None of that is to say that DC is necessarily better, just that a few of your arguments don't hold water.)
It's not the rectification, it's the regulation and filtering. Just because you've ran your 120 VAC through a four-diode bridge doesn't mean that you're ready to pipe the result into your CPU. Not only do you still have a very high voltage, you've still got the sinusoidal waves, they're just all on the same side of ground. You have to regulate the voltage, in this case done (for efficiency) by switching on and off really quickly to obtain an *average* of the target voltage. Once that is done, you have to smooth that out so that you really do have a fairly clean, stable voltage.
That's why when you open a power supply, you see switching FETs (with a heat sink), capacitors, and inductors - with the associated circuitry to make it all work, mostly switching the FETs on and off correctly.
steve
The question is whether 48V DC (which is the most common voltage for this sort of thing) is enough to make your muscles tighten. Under normal situations, it's not.
At this point, some idiot will pipe in with "but it only takes xxx milliamps to activate your muscles, and xxx milliamps to kill you!!!!" Yeah. Volts and amps are not the same, and they're interrelated based on the resistance in question - and unless you're talking about certain moist parts of your body, your body has a pretty fair amount of resistance. As an example from finger-tip to finger-tip, 300 kilohms is the normally quoted standard, although I've measured higher and lower - but the amount varies greatly on the amount of callous, sweat, and other things.
In the end, you can grab a couple of 48V wires in your hand and unless you've got some really moist hands which you've sanded down recently, you won't even be able to feel it. Put them in your mouth, and you'll get a very painful kick, however.
In fact, much to the disbelief of many, even 120VAC, from one hand to the other, barely gives you a tickling/buzzing if you have an average amount of callous (for a man, anyway), and your hands aren't moist. Let them moisten up, though, and things get interesting quickly.
steve
Simple. Because fewer of them are produced, so it's considered a specialty. Just because they're DC doesn't mean that they don't have regulation circuitry on them (at least you had better hope not!). The regulation in an A/C power supply is produced by the million, so the economies of scale kick in. The regulation in your D/C unit doesn't benefit from that sort of quantity.
There's a bit more truth to that than some people realize. One of my acquaintences was a door-gunner in a chopper in Vietnam. He showed me a picture of the side of his chopper, it had the various jeeps, trucks, and other "kills" painted on the side - and a bunch of elephants.
Because the VC sometimes used elephants to carry rockets, if they passed elephants, they were often ordered to fire on them without question of what they were actually used for. A
I could see some profs doing it out of stupidity, but I could see Dan Bernstein doing it entirely out of arrogance...
The cleaning crew in my office had a knack for knocking over my plants, so I bought a cheap USB webcam on ebay for $15 and a USB extension cable, and used some double-sided tape to stick it to the wall. I installed MVC on the machine for motion detection. The problem was solved, no more broken plants.
Well, because the task I desire (avoiding broken plants) is accomplished, I rarely even look at the images - sometimes not for months. But there have been two thefts in the company, and both times, the thief happened to walk down the walkway where my camera is, so I got pictures of them.
Where am I going with this? Just yesterday, I was speaking with the building management, and they brought up the camera - they saw the pictures from one of the thefts, and said that the pictures looked better than most of the very-expensive security systems of other clients in their buildings.
Linux's V4L subsystem makes this sort of thing very easy - all you need is a camera with a V4L-compliant driver, and you're set - there are many packages out there which will take care of the rest. I mentioned that I'm using MVC, which is an old, low-feature application. At some point, when I'm bored, I'm going to switch over to Motion, which is much more full-featured.
steve
Huh. High fidelity hardware for playing audio that's been compressed in a lossy format. Go figure.
Yeah, I know, you got that hoax, too. :-)
Kidneys are a lot different from bones. For kidneys, you need to find a very close match ahead of time, it's not the sort of thing that you can do in the limitted time span in which you can keep the kidney viable. Bone, on the other hand, is just crushed up and sent off to the donor bank, where it can be kept for much longer.
steve
Organ donation is sufficiently in-demand that everything gets used. If you're a full donor, a squadron of doctors come in, and you're more or less left with your skin, muscles, and a skeleton made from PVC sprinkler pipe. Bones, eyes, organs, they take all the good stuff.
I've been in a certain central-american country where they'd kill you for your passport, because they could sell it fo a measly $500.
Dumb terminals? Given that you can stick more and more powerful chips in smaller and smaller devices - with less electrical consumption, it would seem the opposite. Right now, a very good many computing devices are so small that the largest impediment is the *input* - trying to type into the things.
steve
One more thing...
"If I think I've developed a new treatment or prevention of cholera or typhoid, I'm not testing it in Anchorage, AK. I'm going to Manila, Buenos Aires, Rio de Janiero, Cairo, etc. to see if it actually works, once I'm pretty sure that the drug won't kill those it is put into outright. I can't just sneak into a water treatment center and surreptitiously route sewage discharge into the drinking water supply bypassing the filters and Cl2 treatments, etc."
If it were really that important, and it were really isolated to Latin America, then why are they testing it in Norway as well? As cynical as my argument sounds, I've worked for large companies, and I have seen first-hand how these decisions get made. The times when the interest of people at large - even poor people - is a deciding factor in the decision making process are so few that you can't believe it when it happens.
"Hmm... test the drug in the US. 1% chance of a multi-billion dollar lawsuit. The feds keep track of our results. OK, test in Nicaragua, 0% chance of any lawsuit, nobody knows if anything bad happens. What should we do?"
steve
"We have no problem testing new cancer treatments that are on the verge of lethal for "normal" people on terminal cancer patients...why is this any different, then?"
Well, you tell me: Is there a difference between a potentially lethal drug being used on a terminal patient, and a potentially lethal drug (yes, the old one killed some people) being used on someone that's just going to have diarrhea for a few days?
Sure, you could argue that those people would die from dehydration resulting from diarrhea, but having lived in those areas, most of the people which are likely available for the tests also have easy access to pedialyte to take care of that.
Wait until you've seen Latin American cholera wings in hospitals, where the patients are in hammocks with holes cut out for their butts, and buckets underneat them. =)
Yeah, that's a weird subject for an ebola thread.
I once dated a girl who worked in a pathology lab identifying contagious disease samples. When she'd talk about it, she made it sound like she dealt with some pretty serious stuff, so I asked her what she had worked with that day. I don't recall all of the different things, but among them were two different types of ebola. That kind of surprised me. I mean, the 3 or 4 different types of HIV you can expect - with hundreds of millions of people infected all over the world, that's a given, but ebola is vastly less common, and I didn't expect to hear of her dealing with two different types in a single day.
It'll get tested on them for efficacy. Once it's proven effective, then they'll get charged for it. That may sound cynical, but look at what' happened with the recent rotavirus vaccine - since the last rotavirus vaccine was taken off of the market for safety concerns, the new one was tested almost entirely in Latin American countries. The official explanation was that it was needed there more than other places - but the real reasoning is that those locations are easy and cheap to test in, with little or no restrictions, and little (if any) accountability.
steve
It's all a matter of priorities. I'll bet that of those people not willing to pay $25 or $30 for entry-level broadband, a good portion of them spend $50, $60, or more on cable or satellite TV.
steve
Wake me up when^H^H^H^H *if* this ever comes to pass. If I had a gigabyte for every claim I've heard that the next mammoth-sized storage technology (usually optical-based) was just around the corner, I'd already *have* my 1.2 petabytess