Was Thomas Edison Right about DC Power? 545
Declan McCullagh writes "Everyone knows the alternating vs. direct current wars ended with Thomas Edison and Nikola Tesla. But now DC power is being seriously considered for data centers. DC advocates say that plugging servers into AC power is inefficient, and switching to DC cuts down on waste heat and component failure. The University of Florida has even bought 200 DC servers."
Sensationalist, but effectively correct (Score:5, Interesting)
Oh, well, nothing sensationalist about that headline. (*rolls eyes*)
DC advocates say that plugging servers into AC power is inefficient, and switching to DC cuts down on waste heat and component failure.
In this case they're right. With that much hardware that close together, it's easier to treat the entire room as a single device. As the article suggests, this cuts down on waste heat produced by inefficiencies in AC->DC conversion. In fact, it significantly cuts down on the amount of equipment needed in the entire room. The concept can be taken as far as to cutting down to a single power supply per rack.
The amusing part about this is that the resulting racks might look a lot like Big Iron servers with pluggable motherboards.
How Tesla can still make electricity (Score:5, Funny)
Someone has to say it... (Score:3, Funny)
In Soviet Korea, elderly dead people spin YOU!
Re:Someone has to say it... (Score:3, Insightful)
I fail it.
Re:How Tesla can still make electricity (Score:4, Funny)
Re:Sensationalist, but effectively correct (Score:5, Informative)
The article mentions the distribution will be done using 48V for distribution - you will still need DC:DC conversion at the boxes. These DC:DC converters will need to be run at higher current than an AC:DC converter. Higher current can cause more series loss in the system as well leakage losses in the switching supply.
AC:DC converters, as mentioned in the article aren't really that ineffiecient (article itself quotes 90%). AC:DC converters are infact really DC:DC convertors, they just have a rectifier circuit to convert the AC to high volatge DC for DC:DC conversion.
Re:Sensationalist, but effectively correct (Score:5, Informative)
In a standard PC power supply the incoming AC is rectified and stored in a capacitor. Energy only flows into the capacitor when the voltage after the rectifier exceeds that stored in the capacitor. This results in a waveform which departs considerably from a sine wave - no current flows for most of the time while much higher currents than expected flow at the peaks of the half cycles. Electricians interpret this as a bad "power factor" from their experience driving inductive loads where the current lags the voltage by as much as 90 degrees.
Standard PC power supplies are nothing like 90% efficient largely because of this crude rectification of the mains. Compare the rating of your supply in watts with the input voltage multiplied by the input current. These values should all be marked on the case.
Power Factor Corrected (PFC) supplies are available. The better ones use a switch mode circuit to charge the reservoir capacitor through most of the main power cycle, while the less good ones incorporate a capacitor across the mains to buffer the large peaks of current when the input voltage exceeds that stored in the reservoir capacitor.
One advantage of AC is the ease of transforming it to other voltages using transformers and the ease of using it to drive motors especially with multiple phases. In the modern age where switch mode power supplies are cheaper than those using transformers operating at mains frequency this advantage no longer exists. One disadvantage of using DC is the difficulty in switching the stuff off - inductance in the load drives the current straight through an opening switch or fuse creating a nice sustaining arc which is not quenched by the current dropping to zero twice each cycle.
Re:Sensationalist, but effectively correct (Score:3, Interesting)
Second, there are two kinds of power factor - displacement and apparent. A waveform that is not sinusoidal (with high peaks), is said to have a high crest factor. While I am rusty on the terms, one refers to the current and voltage, and the other relates to crest factor.
Re:Sensationalist, but effectively correct (Score:5, Informative)
Re:Sensationalist, but effectively correct (Score:4, Informative)
Most cheap ones use the less good method he referred to(Often called 'passive' vs. 'active' PFC, in PSU literature.)
Why not have separate DC rails for 5V / 12V ? (Score:3, Interesting)
Taking this a bit further, why not have an entire rack power supply that can deliver a rail of 3.3V, 5V, and 12V to each server, thus eliminating the need for a high-current DC-DC converter on the target board? I am excluding things like the exotic voltages for CPU and RAM, but still it is the 12V and 5V rails t
Re:Why not have separate DC rails for 5V / 12V ? (Score:3, Informative)
It's a good idea but it won't fly until DC becomes common in datacenters. And then it won't fly because the datacenters will have all been rigged for 48V.
Re:Sensationalist, but effectively correct (Score:3, Interesting)
Re:Sensationalist, but effectively correct (Score:3, Interesting)
In fact, the biggest problems with today's AC supplies is that the frequency is TOO LOW... this results in the transformers and other AC-AC converters being oversized. Pretty much every switching supply today, including the ones in PCs, chop up either AC or DC into much higher frequencies and this allows smaller components. Avionics have used this for quite a long time, as weight and size savings are crucial!
The
Re:Sensationalist, but effectively correct (Score:5, Informative)
Thus, if you were starting again with an electric grid system, 500 or 1000Hz would be a much better solution.
IAAPD (I am a PSU designer)
1MHz SMPS is nothing fancy these days. In fact they're available even as integrated chips which combine FET switches and the controller into one IC. If you count in point-of-load charge pumps and such, you see up to 3MHz.
Generally speaking, worst offender is high-current FET gate charge which eats up more power than all the other losses combined for synchronous buck transformer (higher DC to lower DC topology, most common type in use probably) Small (1-10uH) inductors are much better behaved in comparison. One reason to use such high frequencies is indeed that you can get smaller inductors and you have less ripple current.. But you're limited by the fet gate charge. Of course, if you're driving some 200W load, you can just say "h*ll with it" and build high power driver circuit to drive your switches, 5% waste heat on switching losses doesn't bring down the house when you can use dirt cheap inductors.
Re:Sensationalist, but effectively correct (Score:3, Informative)
Re:Sensationalist, but effectively correct (Score:5, Interesting)
Thing is how much more efficent is it to have one large ac-dc converter and then smaller dc-dc converters(http://www.nycsubway.org/tech/power/ro
LoL It would be fun to get the DC from old rotary converters for a data center. Big mountain of spinning cast iron with slip rings, commutator, brushes and plenty of copper windings. Put in an old marble switch board with carbon arc breakers , synchro scope, volt/amp meters and knife switches. You then have yourself a turn of the century power system running new millenium computers
Re:Sensationalist, but effectively correct (Score:4, Interesting)
I believe this is SOP in EMP hardened bunkers.
And 4 12V lead acid batteries would be 57.6V
Re:Sensationalist, but effectively correct (Score:5, Interesting)
I at one time had an older NCR ups, a huge old 150 pound honk rated at 1.2 kva, but it could output that 1.2kva for quite a length of time, running these two machines and one of the monitors for about 2 hours one day before I go nervous and did a gracefull shutdown till Allegheny Power managed to roust out a crew into our neighborhood 3 damned days later.
It originally came with a 4 pack of 12 volt gelcell batteries in it of about 12ah each, but when it came into my posession they were toasted.
On checking the float voltage I found it about 2 volts above what I would have called a good float voltage when divided down to a per cell rating, so I knew they'd been overcharged and dried out. I put in 4 18ah motorcycle batteries after setting it down to 52 volts, and boiled them dry in 9 months. I dropped it another volt and replaced them again, this time they lasted about a year before they were bone dry. As I'd rigged the overflow tubes to dump into a small jar of soda, I checked to see if the soda was affected, but it was still as pristine and white as the day I set it up. As that was about $120 a year for batteries, I said to hell with it, stuck a 2 wheeler under it and parked it on the back porch, replaceing it with the same size Belkin, which turned out not to be anywhere near big enough, shutting itself off rather unceremoniously at about 60% of its rated load. I yelled at Belkin and they sent me a much larger unit thats worked for about 4 years now with one battery replacement about 8 months ago.
Idealy I should have been able to run the wet batteries in a stationary environment for 5 to 7 years, and possibly could have if I'd figured out the right float voltage for those batteries.
Perhaps even a fixed trickle of about a milliamp once charged would have worked, but thanks to NCR's habit of burning old docs, I had none on that unit.
I once ran a set of 225ah big truck batteries for 8 years on a standby generator after reducing the trickle charge till there was no more gassing, which was a current of about 5 ma. At the end of 8 years, they would still turn that Cummins 335 hard enough the first cylinder comeing up fired. And the next, second cylinder firing spun it on up enough to kick out the starter, a total elapsed time of maybe 1/4 second from hitting the button and it was only another second to make 1500 rpm and energize the alternator, for a total power outage to on generator elapsed time of about 3.5 seconds. Those 2 batteries would check at about 27.1 volts anytime.
When you've lived where car batteries can freeze and split overnight if not fully charged, one tends to finetune the voltage regulators in the vehicles that must just start, for each battery. I've had batteries that were happy at 15.8 volts without gassing excessively, but in that home made regulator I had strong negative temperature comp too, slopeing down to about 13.8 at 70F, and the next one boiled like crazy at 14 sloping down to 12.4. Each battery has its own 'personality' I guess. Go figure. Yeah, its the old fart again, pontificating a bit about that which he's played with.
--
Cheers, Gene
Re:Sensationalist, but effectively correct (Score:5, Funny)
Why not just deliver the electricity off a truck like everything else in life? This country has gotten so "addicted" to current electricity that we've forgotten that static electricity even exists. A single charged capacitor can supply enough power to run a modern datacenter; the only limitation is capacitance. Say your datacenter runs on 10 kilowatts (that's just a guess) so you need 240 kWh of power a day, or 864,000,000 Joules of energy. Can a capacitor deliver that amount of power?
Sure it can, if it has enough capacitance. Energy storage is 0.5*C*V^2. Say the cap is 1 Farad, and we choose a reasonable charge voltage of 500 kV. [google.com] How much energy is that? 125,000,000,000 Joules! WOW! That will keep you all set for 144.675926 days of continuous uptime! Every couple months, the electricity truck arrives and delivers your charged cap, and you give your spent cap back to the electricity man to be recharged at some high-sulfur coal plant in another state. (That means recycling which will help get the "greens" on board.)
Of course then, you have the nitpickers. "But what about the gasoline for the truck? Isn't that a wasteful means of electricity transmission?" Just use the energy in the caps to run the truck! It's like hydrogen! Hydrogen has already been shown to be politically viable.
Re:Sensationalist, but effectively correct (Score:5, Funny)
Re:Sensationalist, but effectively correct (Score:5, Interesting)
Actually I remember doing some physics problem where I had to calculate the energy density of a simple high voltage paper/oil capacitor charged to near its breakdown voltage. I got an energy density for the cap that was 3% that of gasoline. Chemical fuels are just amazing.
You could run your datacenter off a huge current in a superconducting ring kept near its superconducting transition temperature. As the magnetic field slowly collapses, a circular electric field forms around it. You stick a coil in that field and connect it to your datacenter. Cold magnetic rings can be delivered by (refrigerated nonmagnetic) truck. This scheme is only limited by the current in the ring when it comes off the truck.
You could use a spinning disk. I'm guessing a steel disk spinning almost fast enough to structurally fail and fly apart might have an energy density similar to that of a fully-charged cap. Maybe it's possible to create an ultra-spinnable disk using carbon nanotubes. Then you could spin the disk much faster, and keep your datacenter running longer. I'm too lazy to figure out how fast you can spin a disk like that. But the edge can't go faster than c, or weird relativistic things start happening to the disk. Carbon nanotubes can only get you so far. They could spin the disks up in China, and send them here. But they would have to be careful. If every person in China spun up such a disk at the same time, it might affect the position of the North Star or change the length of the day.
You could just run your datacenter off the 30 tons of TNT.
Re:Sensationalist, but effectively correct (Score:3, Informative)
Re:Sensationalist, but effectively correct (Score:3, Informative)
Re:Sensationalist, but effectively correct (Score:5, Insightful)
I'd love to see how Tesla would have applied his high frequency/voltage engineering to photonics.
Re:Sensationalist, but effectively correct (Score:5, Informative)
Hero Dies Penniless (Score:4, Informative)
Remember the death of Archimedes!
Anyway, the respondant claimed:
Check your history!
Edison died nearly penniless too.
The account which I read described how he ran across some iron-rich sand on a beach, and it gave him the idea to try a new mining technique where the ore would be extracted from non-ore material by dropping the sand past magnets. The idea was a good one (and is still used) but the site he chose to build his iron mine turned out to be almost completely lacking in iron ore. The iron ore in the sand on the beach had apparently washed up from some other source.
Maybe he wouldn't have been desperate enough to try such a risky thing if he had been ALLOWED to sell AC power. I'm sure he could see the advantages of AC for power transmission but Edison didn't have the patents for that, and you can bet that Westinghouse wasn't going to license the technology at a reasonable price to their chief competitor.
So Tesla got ripped off by Westinghouse because he wasn't business savvy and they got ownership of the patents. Then Edison, even though he was somewhat business savvy, got shut out by Westinghouse because they owned the patents. In both cases, patent law helped business-people who didn't invent anything get rich while the real inventors lost out. Shouldn't we remember that the patent system was set up in order to encourage invention?
Re:edison vs. ac (Score:3, Interesting)
Tesla 1, Edison 0.
Re:Sensationalist, but effectively correct (Score:5, Informative)
In particular, "Increased stability of power systems" is certainly something that individuals in the Northeastern US and London may be interested in.
Of course, AC still has its uses, but the chart is now thought to be:
really long distance -> HVDC
long distance -> AC
short distance -> DC
Re:Sensationalist, but effectively correct (Score:4, Insightful)
Re:Sensationalist, but effectively correct (Score:3, Interesting)
Well that's more about current - most 48v supplies are capable of delivering quite high currents (lower voltage == higher current for the same wattage). So whacking a big conductor across the circuit like a screwdriver means you'll get a massive current compared to most high voltage supplies (which would've blown a fuse for similar currents).
OTOH your body is not a great conductor so the fact there is a higher current available doesn't mat
Re:Sensationalist, but effectively correct (Score:3, Interesting)
This is not true anymore. Most new long-distance lines being built around here are DC. DC is far easier to regulate, so it helps mitigate the risks of grid breakdowns. It is also more efficient to transform DC.
New Power System (Score:5, Funny)
Re:New Power System (Score:3, Funny)
AC/DA (Score:2)
no, dude, that's a band. (Score:2)
Number one with a bullet, I'm a power pack! (Score:3, Funny)
Yes, I'm back in black!
[tap tap]
Hey, who turned off the microphone?
Was Edison right? (Score:2)
No, he was having a pissing match with Tesla.
Re:Was Edison right? (Score:2, Insightful)
Just think, if he'd settled with Tesla back then, today they could be sending people to be killed on the Edison Chair.
Re:Was Edison right? (Score:4, Funny)
Old news, for Verizon WA (Score:3, Informative)
Phone companies are all DC powered (Score:5, Informative)
This goes back to the telephone talk battery, which is -48 V DC. That powered the phones via old cord switchboards, and was the voltage of electromechanical (stepper, and later crossbar) switches, which basically used relays. Electronic gear was then designed to run on the same power plant. A telephone building has a big bank of batteries, powered by multiple "rectifiers" (DC supplies) which, btw, are normally engineered to not run over 40% of load. (That way they can still run the systems and recharge the batteries when one of them is kaput.)
If you then put anything else into one of their buildings, the Network Equipment Building Standards (NEBS), which are Telcordia documents that practically carry the force of law, dictate that equipment be DC powered. Among other things -- NEBS gear has to meet the brick schytthaus test. (Sun Netras and many Cisco routers meet NEBS. Your basic rack server doesn't. And aluminum racks are STRICTLY forbidden; it has to be steel.)
So because of the talk voltage on analog phones, lots of computing equipment is engineered for -48 V DC power. Sort of like the legend (I know, that one is not really true) about the railroad track gauge being based on Roman chariots. But in this case it's surprisingly effective.
Telco, telegraph, computer, and deejay... (Score:5, Informative)
And the reason it's negative with respect to ground goes all the way back to the telegraph system: Western Union initially ran bipolar lines and noticed that the positive ones corroded much faster. Sodium ions (from dissolved salt) are negative, and thus repelled from lines that're also negative. The whole phone system was built with positive ground because of this, and it's saved incalculable maintenance costs. It does tend to mess with people's heads the first time, if they're used to negative ground systems, but you get over it quickly. (A number of traditions use blue for "hot" and black for ground/return, to help escape your "red equals positive" association.)
DC power as used by telcos is also always redundant. There's an A-side and a B-side for everything, and the cables are sized so that the entire load can run from just one side. This leads to some very fat copper, which is cheap compared to downtime. You don't achieve five-nines reliability with a system that contains single points of failure!
Now, about rack-mounting: This was also invented by the telcos, originally in a very wide (40-inch?) format, for the panelboards and Strowger switches. Some of the old crossbar equipment is still in those huge racks, but the 23-inch width is infinitely more common now. All telco equipment is mid-mounted, with the ears approximately in the center of gravity on the shelf, so the force on the screws is shear. There's no torsion on the mounting flange unless you step on the front or back of the shelf. Cooling is always convective bottom-to-top, or occasionally front-to-back with fans. This leads to a "cool" front aisle and a "warm" back aisle between alternating rows of equipment.
Now, the pro audio industry borrowed the rackmount idea fairly early on, but they were mostly mounting control panels and mixers, which are very shallow, so flush-mounting made sense. They also changed the every-inch Western Electric mounting holes to an alternating-spaces "EIA" standard, and narrowed the rack from 23 to 19 inches.
Somewhere along the line, an absolute idiot decided that computers should be rackmounted, but they should be 19 inches wide, flush-mounted, and use EIA hole patterns. I'm sure this has something to do with mainframe legacy getting perverted by peecee people. The current mishmosh of mounting standards (19" vs 23", two-post versus four-post, flush versus mid, inch versus RU, front-cable versus rear-cable) is what every datacenter tech deals with on a daily basis. Throw overhead racks versus raised-floor cabling into the mix, and you've got a recipe for frustration!
If you're familiar with the concept of "blade servers", where common components are separate from processor resources in the shelf, congratulations. Telco hardware has been built like this since the invention of the circuit board. Actually, the concept of replacable plug-in units goes back before that, but it got vastly easier with printed wiring boards and card-edge connectors in the sixties. Most of the "good ideas" in serious computing circles are actually century-old ideas in the telco industry. Spend a week shadowing a central office tech before you design a datacenter, please!
Also consider: If your datacenter is already built for DC, throw some solar photovoltaic panels on the roof. Inverters are a large part of most PV systems' expense, and you can skip that part. Why not start offsetting your grid demand now?
Also also: Edison was flat-out wrong about DC. The modern switching power supplies that make DC transmission lines practical didn't exist in his day. Besides, long-distance power transmission is an entirely other discussion.
Re:Telco, telegraph, computer, and deejay... (Score:3, Informative)
Sodium ions from a salt are definitely not negative. Sodium like many other akalai metals tends to lose its outermost electron and form a positive ion. I think you'd have a hard time getting sodium to pick up an extra electron.
It makes the rest of the explanation a bit hard to swallow.
Re:Telco, telegraph, computer, and deejay... (Score:4, Informative)
They were both right...and wrong... (Score:5, Insightful)
For moving power over long distances, AC is king. But for short distances with most modern electronics, DC would win. The first thing a desktop system or server does with AC is converts it to DC. So if you have a number of machines all in the same room, why not do the conversion in one spot, and eliminate the redundancy in every machine.
Would it benefit the average user with one or two machines? Not at all. But for a major center with many machines in the same room, I can see quite a bit of benefit with going with DC.
Re:They were both right...and wrong... (Score:5, Funny)
Nope, but I've put in screws with a hammer, even when I had a screwdriver on hand.
Re:They were both right...and wrong... (Score:2)
Re:They were both right...and wrong... (Score:5, Informative)
Nope. For the longest-distance transmission lines, you see DC being used. There comes a point when the capacitive losses you get from using AC encourage you to switch to DC, and for lines of several hundred miles, you start seeing DC transmission lines.
Re:They were both right...and wrong... (Score:4, Informative)
Re:They were both right...and wrong... (Score:2)
Re:They were both right...and wrong... (Score:3, Informative)
HVDC is actually FAR better for long line power distribution due to ACs inductive line losses... IIRC DC _is_ used some places. (California)
The downside is that AC requires only a series of transformers to step it down to various levels for local power distribution--- Makes for a relatively straightfoward infrastructure.
DC for all practicle purposed MUST be converted to AC for this purpose, via big honkin
Electrocuting an Elephant (Score:5, Interesting)
In one instance, he even electrocuted an elephant...
During the construction of Luna Park on Coney island, an elephant used as a beast of burden went out of control and killed a couple of people. Topsy, as she was called, was condemned to death. However, there was a wee bit of a problem. Elephants aren't the easiest critters to kill. What happens if you walk up and fire a shotgun at it's head, only just to piss if off? They do have rather thick hides, and we are talking about a homicidal elephant the size of a couple SUV's here. There weren't any cliffs handy to stampede poor Topsy off of, and I doubt dynamite was ever seriously considered. Edison, being the generous person he was, gladly volunteered to execute the elephant with AC current and filmed the whole thing. He showed the resulting film, "Electrocuting an Elephant" (1903) publically on many occasions. It is quite probable that many a cat and dog escaped a crispy fate thanks to this film. If you decide to track down a copy of "Electrocuting an Elephant" today, please be warned that it's a rather gruesome little piece of history, and is not for the faint of heart, or SPCA members.
Re:Electrocuting an Elephant (Score:4, Informative)
Really though it's not that good, it only gets 4.2 stars at imdb [imdb.com].
-s
Thanks for reminding everyone (Score:3, Interesting)
It's true only in a pretty restricted sense (Score:3, Informative)
How come there is no real difference? Because both modern AC and modern DC supplies start out by converting the power to high frequency AC (on the order of several kHz), and operate on that. That's what you actually want as input, if anything.
The article states:
In other words, the DC supplies they use are more efficient than standard AC supplies, which are the cheap crap and notoriously inefficient.
Re:It's true only in a pretty restricted sense (Score:2)
"DC with pulse with modulation" -- sounds like an alternating current to me. Yes, it's typically a very messed up square wave (due to all kinds of filtering) rather than a sinusoid pure sinusoid, but spectrally, it's just noisy AC.
Re:It's true only in a pretty restricted sense (Score:3, Informative)
DC->DC converters are basicly AC power supplies. They pulse the DC current up to several hundred kHz, using an inductor, and convert it down/up on the other side. They're very efficient, although somewhat costly.
Tesla strikes back with wireless power! (Score:2)
I'm pretty sure I'm just joking about that idea...
Re:Tesla strikes back with wireless power! (Score:3, Informative)
Eh? (Score:2)
Is this not the entire point of the PS? To convert AC to DC?
So basically all these new DC computers would be is a computer that relies on yet another source to convert AC to DC then still requires some sort of internal component to convert THAT DC to the correct voltage for the various devices within the computer itself?
Re:Eh? (Score:2)
It will make the servers smaller, save space by consolidating power supplies into one unit, save lots of power by having one large power supply that's more efficient than small ones, and make it way easier to remove waste heat by concentrating most of it into one big unit.
Re:Eh? (Score:2)
Wow (Score:2)
Perfect (Score:5, Funny)
Brilliant!
this is news? (Score:5, Informative)
Re:this is news? (Score:3)
Phone company example (Score:2)
Edison was wrong (Score:2)
No (Score:3, Informative)
I've seen houses wired with 12V DC from mini hydro and solar - but in those cases it was a long way to the nearest transmission wire and would cost a fortune to get mains power onto the site.
Telcos have run on DC for decades (Score:2)
To Westinghouse (Score:2, Informative)
Incidentally, that's how the electric chair came about:
[Edison]AC is dangerous! Just watch what happens to these various animals when I close this circuit!
Edison electrocutes some horses
[US_Gov]Ooooo... I'll bet that works on people too!
US_Gov introduces new grisly method of executions, while disregarding the main point of Edison's demonst
DC for a building is good, but (Score:2)
-48v Power (Score:2)
Misinformation in article (Score:5, Informative)
That said, if space and cooling are an issue it might well make engineering sense to get the transformers, capacitors, and rectifiers out of the computer boxes. Big 5v/12v power busses wouldn't even need to be insulated. So while the reporter badly mangled the story, the engineering sounds reasonable to me.
Yes he was? (Score:2)
However, if you enjoy distributed power, no, no he certainly was not.
Not over! (Score:2)
Re:Not over! (Score:3, Interesting)
Yep. My dad was the building superintendent of a church on 96th St. in NYC in the early 70s. The church building has a DC mains supply - mostly to run elevator and fan motors, but some of the outlets in the building were DC, and were identical enough to normal AC outlets that you could plug a regular plug into them. Well, before he knew better, my dad plugged an old TV into a DC outlet. Transformers don't take DC input very well - fireworks en
DC vs AC in data center is about efficiency/heat (Score:3, Informative)
Here is the Answer (Score:2)
-Angus Young "AC/DC"
If you don't get this post, I deserve -Troll, -Offtopic and -Flambait karma from each and every one of you.
The trouble with 48VDC land (Score:3, Insightful)
What Rackable is really pushing is a system where AC to 48VDC conversion takes place in a unit at the top of the rack, and 48VDC is local to the rack. That, at least, simplifies the cable management.
One big advantage of 120/240VAC power distribution using US standards is that the connectors are standardized and reasonably idiot-proof. That is, if you can plug it in, you won't overload the power cord or the connector, and if you overload the branch circuit, a breaker will trip. Outlet strips have circuit breakers, so you can't overload the cord to the outlet strip without a breaker trip. There are NEMA standard power plugs [leviton.com] for 15A, 20A, and 30A circuits, 120/240VAC, and single and three phase configurations. All this is standardized nationally and enforced by the National Electrical Code.
In contrast, there are no simple standards for 48VDC. Most 48VDC gear has big screw terminals. There are no standard plugs and sockets. 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. None of this is all that hard if you have basic power technician skills, but you can't just go casually plugging stuff in.
Although, since the development of the low-cost clamp-around DC ammeter, things have become easier in the DC world.
UPS? Of course DC is more efficient! (Score:3, Insightful)
Contrast this with a properly designed DC system a la old-school telco: The same front-end of the UPS is used, with a 10% loss converting AC to battery voltage. Then you run that into DC supplies that, with modern electronics, are going to be doing a lot better than an AC supply, so let's say 5% loss. That puts you at better than 85% efficiency.
The critics cited in the article are actually probably not far off in calling the Rackable solution over-hyped, if you only take into account the isolated-rack design. Rackable puts 2-3U of beefy redundant supplies at the top of the rack and does DC to the servers. Efficiency-wise this is only fractionally better than a bazillion AC supplies, and quite possibly dead even because of the DC->DC losses in each server on top of the AC->DC->AC->DC setup implicit with AC-based UPS systems. However, AFAICT from a glance at their site, Rackable's systems are designed to drop right into existing DC datacenters, which eliminates the AC supplies at the top and the DC->AC->DC stages.
The issue is what kind of infrastructure is needed to feed the selected DC voltage (which is going to be -48VDC) into the racks with the lowest bus losses, but this is someone I would expect is either a) already solved by the decades-old telco industry, or b) going to be solved in at the appropriate 384-cores-and-100TB-per-7ft-rack scale RSN, by "the market".
I know that if I were in the position of designing a big datacenter right now, I would be looking very hard at DC systems.
They're conflating several things in the article. (Score:5, Informative)
First off: Digital electronics generally requires several voltages. And they're all low, requiring high currents, massive conductors, and local filtering and regulation. So even if you're providing DC power from outside the room, you'll have a switching power supply (or several) in each piece of equipment to convert whatever the rough DC power is to whatever you need, smooth it, and regulate it.
But while some electronic devices use a common switcher to generate all the voltages with one conversion step, others use a "roughing" supply and a bunch of local supplies. Part of that is to get better regulation - part is because the roughing supply must run from 60 (or 50 or whatever) Hz and thus requires big caps to tide you over the low part of the cycles - caps you don't want taking up space near the components.
If you're going to do it in two stages anyhow, you can put your roughing supply OUTSIDE the room and only have the final supplies inside. The roughing supply has a lot of heat dissipation so you save a bunch on your cooling.
Second: There are two standards for power distribution in electronics rooms:
- Your local power line stuff. (120/240/480/208-3-phase in the US)
- The telco standard: x2-redunant 48V DC.
A lot of equipment - especially networking equipment - is manufactured for sale to tellcos and other operations that use the standard. They might have initially used it because some of their equipment was co-located in tellco sites, where only 2x48VDC is available - and they got a quantity discount for buying a bunch of the same stuff and went to 48V for their own sites. Or they might use it because it's MUCH simpler to do backup power with floating batteries and century-old technology than with a building-sized UPS. (Note that a UPS CAUSES at least one outage when first installed and on the averate at least one more within the first year of operation from some malfunction. And a UPS dissipates more power than a roughing power supply or a battery charger.)
But the standard for 48VDC is REDUNDANT 48VDC supplies, with the equipment only requiring one (and typically doing "cutover" with diodes B-) ). With the equipment already set up for redundant supplies it's not a lot of cost or work to wire both sides and put in two 48V feeds to the equipment room. (Four diodes are a LOT cheaper than a pair of 120V roughing power supplies at each box, too.) So of course the users of such equipment normally give it dual supplies. (Even if it's a single rack and so they just put two roughing supplies in the rack fed from two different 120V feeds.)
The result is that all the equipment has redundant power supply, and keeps operating glitch-free through a number of kinds of partial outages - AND power supply repair and replacement. This is what's responsible for much of the claimed increase in reliability.
The whole Edison/Tesla DC/AC war had to do with the economics of CROSS-COUNTRY power transmission. AC beat DC there because a century or more ago it was virtually impossible to jack DC voltages up to levels suitable for long-distance transmission and back down to levels safe for distribution within houses, while AC could do that easily and efficiently. So Westinghouse/Tesla could ship cheap power from Niagra Falls to New York City while Edison had to build fuel-burning power plants IN the city. It has essentially nothing to do with shipping the power around within a single building.
right and wrong (Score:5, Informative)
Ever seen a telco rack? Everything runs on -48VDC. Everything. A telco rack always includes a couple of DC power supplies, and all the equipment just ties in to a common DC bus. The best part of all: the UPS simply consists of four "car batteries" (not exactly, but you get the idea) wired in series and tied directly into the bus! No pesky inverters to deal with.
The telecom industry has been doing it this way for decades. It's about time the computer industry got on board.
Re:right and wrong (Score:3, Informative)
Switching power supplies that are optimized for a small range of load conditions can achieve 95% or better efficiency. Most computer power supplies (built for a wide range of load conditions and voltages) are about 85-90% efficient now. Simple rectification and regulation through a linear regulator loses various amount of power through heat, depending on the load, but varies from 50-70% efficiency in a good design. This is what most wall-wart tra
This isnt the same AC vs DC debate. (Score:3, Informative)
TTL logic has to run on DC, so you have to convert the supplied AC to DC. This is just recognizing that instead of converting it individually in each of dozens or hundreds (or more) machines, that it is more efficient and reliable to have one (and perhaps a redundant standby) converter providing DC to the same machines.
AC vs. DC for power transmission (Score:3, Informative)
Losses are especially bad in AC transmission lines when the power factor is not correct, because while currents which are out-of-phase with the generated voltage waveform are expressed using imaginary numbers, in fact they are very real currents, and they cause increased heating losses in the transmission line. So the power companies switch large capacitors in and out of the circuit to try to keep the current and voltage in phase. (And they would appreciate if every device on the grid was power-factor-corrected, but this doesn't happen, mostly because motors are inherently inductive, and motors are the largest consumer of electricity. Sometimes they at least manage to persuade large industrial customers to manage their own capacitor bank, to correct for the inductance of their own motors, and give them a discount in exchange.)
Re:Westinghouse (Score:2)
Re:Westinghouse (Score:3, Interesting)
Re:Antistropic Magnetic Fields (Score:3, Insightful)
"Anisotropic Magnetic Field" has to be the worst offense in terms of technobabble i have seen recently.
Newsflash: there are no magnetic monopoles, so EVERY magnetic field is anisotropic...
Re:Antistropic Magnetic Fields (Score:3, Insightful)
It doesn't seem too surprising that AC power would produce an anisotropic field since the current keeps switching, so the magnetic field should be switching direction also. I suppose this would make the magnetic field from a DC current isotropic (invariant with direction, or I suppose in this context constant orientation), but I don't really see why either would be an issue (since you referred to Maxtor, I assume the issue was something that had to do with hard drives). A
Re:Uhh... (Score:2)
Re:Uhh... (Score:2)
Re:Uhh... (Score:2)
Re:Uhh... (Score:2, Funny)
Re:Uhh... (Score:2, Informative)
Re:No, Thomas Edison was wrong (Score:2)
DC-DC converters work by chopping the input power into high frequency AC, which require smaller magnetics than typical line frequencies, which were selected for the benefit of early rotating machinery. If you don'
Re:Copper bus bars?? (Score:2)
Re:"220?" (Score:3, Funny)
Re:AC versus DC (Score:3, Informative)