Forgot your password?
typodupeerror
Hardware Hacking Hardware

DC Power distribution - Nix the Transformers? 180

Posted by Cliff
from the we-call-'em-wall-warts dept.
MailtoDelete asks: "I have most of my electronic gear plugged into a couple power strips hanging off a UPS. Most of these devices have big block-type transformers which, besides being bulky, are a bit of an eyesore. I have been trying to find a product out there somewhere that would allow me to have one central transformer that would distribute DC power at variable voltages, depending on what devices I wish to plug into it (think one AC input and 9 or so DC outputs individually adjustable). I found this device that resembles what I have in mind, but it does not have sufficient output for my router, switches, and various other devices. Is there a product on the market already that would do this? Can I build one with my marginally above average soldering and electrical schematic skills? Have any of you found a better way to eliminate these blocky plug-hogs?"
This discussion has been archived. No new comments can be posted.

DC Power distribution - Nix the Transformers?

Comments Filter:
  • by Jeremiah Cornelius (137) on Tuesday March 01, 2005 @06:39PM (#11817488) Homepage Journal
    "This is Reddy Kilowatt, reporting a dangerous, overloaded octopus connection in sector five!"
    • Re:Hardware Wars (Score:5, Informative)

      by thegrassyknowl (762218) on Tuesday March 01, 2005 @06:46PM (#11817561)
      Wasn't it Thomas Edison who tried to prove that Tesla's 3 phase AC power distribution was dangerous by electrocuting frogs with it and showing how they thrash about vioilently before they died?

      *tongue firmly in cheek*
      Perhaps DC power distribution is the best after all.

      That said, you could easily build a device to power all those said gizmos. You'd really need a quite large multi-tap transformer with appropriate ratings, and a set of voltage regulators for the various voltages... 5, 6, 9, 12, 13.8, possibly a couple of adjustable ones for those pesky items that insist on odd voltages.

      I had a similar (homebuilt) device with 6 outputs, all individually regulated.

      This is a good excuse for a PIC-type project to set the ouput for each port... It could adjust the regulator to get the right voltage and also toggle relays for each port to get the right transformer tap (so as to avoid dissipating too much energy in the little regulators) for any given voltage.
      • That said, you could easily build a device to power all those said gizmos. You'd really need a quite large multi-tap transformer with appropriate ratings, and a set of voltage regulators for the various voltages... 5, 6, 9, 12, 13.8, possibly a couple of adjustable ones for those pesky items that insist on odd voltages.

        I believe such devices are called "power supplies". You can get hobby power supplies up to some pretty insane voltages with lots of different output voltages (and you can use voltage regul
      • Frogs. Sheep. Horses. Elephants. People.
      • Re:Hardware Wars (Score:5, Interesting)

        by snorklewacker (836663) on Tuesday March 01, 2005 @06:58PM (#11817698)
        Wasn't it Thomas Edison who tried to prove that Tesla's 3 phase AC power distribution was dangerous by electrocuting frogs with it and showing how they thrash about vioilently before they died?

        Not Tesla, Westinghouse. The whole frog twitching thing was a sideshow trick when electricity was first discovered, and could be done with DC. Edison went all the way up to electrocuting horses, and advertised it could be done on people with "Westinghouse's Electric Chair". He thought people would be horrified. In the last bit of irony, several states loved it and actually ordered these things, using them for capital punishment for many decades afterward.

        Edison may have been quite an inventor, but he was rather a ruthless man not above gross distortions and character assassination.
        • Re:Hardware Wars (Score:5, Informative)

          by alienw (585907) <alienw...slashdot@@@gmail...com> on Tuesday March 01, 2005 @07:05PM (#11817777)
          He was as much an inventor as Bill Gates is a programmer. Above all, he was a businessman. Most of "his" inventions were actually created by the people working for him.
          • He was as much an inventor as Bill Gates is a programmer.

            Interesting analogy. I wonder if Bill Gates would electrocute a penguin to prove WIndows is more secure than Linux.
          • He was as much an inventor as Bill Gates is a programmer. Above all, he was a businessman. Most of "his" inventions were actually created by the people working for him.

            Edison was born in 1847, by 1870, he had a national reputation for his work in telegraphy, and by 1879 he had been granted 170 patents. Edison's Patents 1868-1879 [rutgers.edu]

            • Re:Hardware Wars (Score:3, Informative)

              by Grab (126025)
              Alert - getting patents and getting a reputation for stuff your lab does do not mean you're so damn hot. In fact, it doesn't even mean you invented it - witness the *many* times Edison ripped stuff straight off other people. Nor does it mean the inventions are any damn good - Edison was quite happy to use publicity, lawsuits and outright lies to promote his stuff and crap on other people's stuff.

              So a lot like Billy Gates and his organisation, in fact...

              Grab.
          • True dat. Tesla was the real inventor back then, which is why we wound up using AC polyphase instead of Edison & Co.'s DC system. Many bitter exchanges between the two camps, with Edison's side performing their famous "Westinghousing" of dogs and such to prove AC's danger. (The name comes from the fact that George Westinghouse had bought/leased/stolen Tesla's patents for AC production/distribution.) Fortunately the science of Economics won out when it was shown that AC would be far cheaper to produce an
        • Nit Alert -

          Tesla worked for Westinghouse
        • The frog twitching went way back, to Galvani, I believe, and inspired Mary Shelley's Frankenstein.

          Edison went all the way up to electrocuting horses

          He also did an elephant. [roadsideamerica.com]
        • Edison may have been quite an inventor, but he was rather a ruthless man not above gross distortions and character assassination.

          ...and electrocuting horses for kicks. Jeebus!

        • Edison went all the way up to electrocuting horses, and advertised it could be done on people with "Westinghouse's Electric Chair".

          Edison even electrocuted an elephant to prove how "Dangerous" AC power was.

          Even so, the fact they AC didn't require substations every 3 miles made it more practical for widespread distribution.

          LK
      • Sure - Edison lost to Westinghouse.

        The issue was how to get energy from the Niagra falls. People discussed compressed air, DC, and AC, and Westinghouse showed that by jacking up the voltage, he could use a small(er) guage wire.

        DC is fine, but at the time, there wasn't any good way to increase the voltage.

        Ironically, 50-60 hertz is fairly slow, and not the most effeceint frequency.

        Now - with semiconductors, transformers, and high speed transformers (switching Power supplies) we could do better - but ther
        • One who knows about such things was discussing this with me just the other day. It was in respect to digital communications. There is RS-232 which is always +/- 5, 10, 12, or 15 volts (depending on what kind of power is in the system) and measures voltage drop for the bits. The limitation is that you can only send 15 volts so many feet of cable before the gauge gets too low (or high, depending on how you look at it) to be practical--the cable just gets too thick for reliable communications.

          The alternati
      • I know he electrocuted elephants.

        Here's a video:
        http://homepage.mac.com/zav/Elephant.mov
  • by Anonymous Coward
    The device you're looking for is a power supply -- they're not just for computers. Just get one that will put out enough amps for peak demand on all your devices, mount some molex connectors onto a board of some sort, color code them by their voltages, and stick the whole shebang in an old PC case.

    Of course you'll have moved from a bunch of smaller blocky transformers to a large honking thing with a fan, you'll have to wire up your own cables. Watch that you don't overload it, or you let out the magic sm
  • by hoggoth (414195) on Tuesday March 01, 2005 @06:50PM (#11817613) Journal
    Forget replacing wall warts with one very large wall wart.

    Where is my wireless power?!

    We wont be truly untethered until we have wireless power.

    (Not entirely kidding. Is there any safe way to deliver wireless power? Or am I just asking to turn my house into a very large Microwave Oven?)
    • How about batteries?
      • by hoggoth (414195) on Tuesday March 01, 2005 @06:59PM (#11817715) Journal
        Batteries? Great.
        I already HAVE a battery for my server. It weighs 50 lbs and is as big as a shoe box. It's called a UPS and lasts less than 30 minutes.

        I don't think that's the answer to everything electrical in my entire house.

        The battery for the Fridge will be bigger than the fridge :-)
    • Re:Wireless power! (Score:3, Insightful)

      by raider_red (156642)
      We could probably set up a wireless power system, but you can forget about the "safe" part. You could also give up any hope of ever having children.

    • Re:Wireless power! (Score:3, Informative)

      by fm6 (162816)
      Already been tried! [prometheus.al.ru]
    • There exist some forms of wireless power already, basically in two categories; port-a-power (batteries, capacitors, etc), and transmittable power (microwave/RF, magnetic inductance, lasers).

      You can transmit power a short way without wires using magnetic inductance, like in some electric toothbrush rechargers (you just set them in the cradle and they charge), but you need to be very close to the originating magnetic field. You can also use radio frequency to transmit some power, which is how some RFID tags
      • Other than that, I don't think there are many other options.

        Tesla would have disagreed. Check into his Colorado Springs research sometime... also in the unveiling of the Niagra power plant (which he helped design) he pissed alot of investors off by giving a speech about how wired power is obsolete.

        His idea was basically to use the Earth/ionosphere as a huge capacitor which could be charged from anywhere and consumed from anywhere with around 70% efficiency.
  • No. (Score:2, Interesting)

    by Short Circuit (52384) *
    Can I build one with my marginally above average soldering and electrical schematic skills? Have any of you found a better way to eliminate these blocky plug-hogs?

    Sure, but you'd need a transformer with multiple sets of windings, one for each voltage you want to put out, and one for your line input... But you're not likely to find one that fits your exact needs.

    Nevermind concerns about drawing too much current from your device, and failing gracefully.
    • Re:No. (Score:3, Insightful)

      by alienw (585907)
      You don't need a winding for every voltage. One winding with multiple taps should work just fine. As far as current draw: a circuit breaker or fuse on the primary will do the job.
      • I don't get all the talk about multiple windings/taps for each voltage. Why not one voltage (the highest required by the system) with oodles of current behind it, then simply regulate it down for each voltage. I think I remember that EIN max for an LM7805 (TO220 package) regulator is 12 volts or more. Just mount all your regulators on a big heatsink, regulate and filter from the common rail, and you're in business.

        • Re:No. (Score:3, Informative)

          by Hank Reardon (534417)

          Why not one voltage (the highest required by the system) with oodles of current behind it, then simply regulate it down for each voltage.

          I seem to remember from my electronics courses, many years ago that the LM series of regulators work similarly to a resistor; they dissipate over-voltages as heat. Now, while you do need some overhead voltage for the regulator to function properly, too much and you'll fry the thing. The TO220 package also had a maximum current rating of somewhere in the neighborhood of

          • yeah - this is all a big - why don't we use DC argument.

            The answer is this:

            Transformers are effecient

            Resisters / semiconductors are either A. ineffecient, or B. Noisy - as in producing square waves.

            If you don't understand B. then focus on A. like a mantra. scratch it into your screen with a key. Have your mother sing it to you as you drift off to sleep.

            AIK

          • I'm a little late with this, but:

            Linear regulators will work like a resistor, as you say, getting hot and inefficient.

            Switching regulators, OTOH, are much more efficient. Basically, they turn the power on and off very quickly and do so to charge a capacitor. They monitor the voltage on the capacitor and adjust the duty cycle to make sure that the voltage on the capacitor stays where you want it to... if the load on the regulator increases, the capacitor will drain quicker and will need a higher duty cyc
        • by Bishop (4500)
          Using voltage regulators as you suggest is very inefficient. For example if you have a device which draws 250mA at 2.5V you are burning 2.37watts (9.5V * .25A) just to supply 0.62watts. That sucks.
        • by alienw (585907)
          You don't want to drop more voltage than you have to across the regulator. If you have 2 amps at 12V coming in, and 2 amps at 2V going out, you are losing 10 * 2 = 20W on that regulator. That's one HOT regulator. That will require a big heatsink and fan -- not to mention it's really inefficient.
          • The Ask Slashdotter is looking for convenience, not power efficiency.

            And as to a previous poster's reference to rectifying/filtering, well, I took for granted that that was understood.

            The original 7805 reference was a "jumping off place" for searching for suitable regulators. Newer versions come in a variety of packages, from tiny surface-mount to TO3 and larger packages. The big ones with a proper heatsink and mounting can dissipate a lot of heat, and the new designs are a lot more forgiving with large i
    • First, why would you need a multitap transformer to output different voltages? Simply set up a switch-mode power supply, like in a computer PSU: use a bridge circut, transformer and a couple of filter caps to get a few hundred DC volts, then use a transistor to pulse the output at a few hundred kilohertz. Output voltage equals duty cycle times input voltage; run it through a simple filter and you're done.

      As for overcurrent protection, there's been a device around as long as mankind has harnessed electrici

      • by alienw (585907)
        I take it you know very little about switching power supplies. Just so you know, they are rather complex and require a solid background in control theory, electromagnetics, and electronics to design and build. The EMI issues are challenging, and it's not a simple matter of applying a fixed duty cycle (since the voltage is proportional to the load).
        • by chthon (580889)

          Yeah, and he should not forget the fire extinguisher as he tries to build such a thing.

        • Re:No. (Score:5, Interesting)

          by harrkev (623093) <kfmsd@NosPam.harrelsonfamily.org> on Wednesday March 02, 2005 @10:05AM (#11822653) Homepage
          What the parent poster said! I design boards for a living (along with doing some VHDL stuff in FPGAs). My expertise is digital, but I do have an Electrical Engineering degree -- and I could not properly design a switching power supply -- at least that what what I was told by my power supply guy after I tried laying out his circuit on the board.

          I have also tried my own experiments in this area. I had an old Celeron 466 system, and my scanner, speakers, and force-feedback joystick all ran off of 12V. So I hooked up my computer's 12V rail to some connectors on the back (with caps attached to help reduce noise). Here were the results:

          1) Scanner - worked like a charm. But you had to plug in the power while the computer was off. Otherwise, the whole computer would reset iself. This was my one success.

          2) Speakers - Worked, but you could hear the hard drive heads moving. The amount of noise was too much, so I had to go back to a separate wall-wart.

          3) Force Feedback Joystick - This was a Microsoft model, which connected to the Game port (not a USB model). Apparently, this stick did not like sharing its power ground with its signal ground. The computer shut off instantly, and the joystick was fried. This goes to show that you need to have FLOATING power supplies in any scheme like this.

          In the end, my grand experiment was not really worth it, as I spend a couple of hours soldering and drilling, and all I did was to remove one power supply.

          But if somebody DID make a power supply box that had the following: 3x12V, 2x9V, 2x6V, 3x5V, all at 2A each, and they could do it for under $100, then I would be very tempted to buy one. Unfortunately, I suspect that this type of supply would go for a lot mroe than $100.
  • Liberator (Score:3, Interesting)

    by Zakir (849137) on Tuesday March 01, 2005 @07:00PM (#11817723) Homepage
    This doesn't fit what you're looking for exactly, but it fill help the problem of having to plug wall worts into your power strip. It is basically a 12 inch extension cord that plugs into your power strip so that the transformer doesn't take more than one outlet up.
    "Get full use of your power strips and UPS outlets with this premium power cable from Cables Unlimited! Just plug this cable between your bulky power adapter and any unused outlet and this revolutionary designed space saving cable acts as a 1' extension, giving you a little extra length to get into hard to reach places."
    http://www.tigerdirect.com/applications/ SearchTool s/item-details.asp?EdpNo=392776&CatId=1284
  • Your link. (Score:3, Interesting)

    by Murphy Murph (833008) <sealab.murphy@gmail.com> on Tuesday March 01, 2005 @07:02PM (#11817748) Journal
    The link you provided to the "ELK-PD9 Power Distribution Module" is telling.

    Do you understand why devices such as that are used?

    A power distribution module such as that is used to power devices like remote video cameras, remote security sensors, and other remote monitoring devices.

    Key word being remote.

    Devices such as these use structured wiring with data (video) and power coming in to the device over one cable bundle. This is done for ease of installation to the remote site, and because the security camera on your neighborhood Target store being 100 feet from the nearest power outlet makes a wall-wart is unfeasible not unsightly.
  • Why? (Score:5, Insightful)

    by Guspaz (556486) on Tuesday March 01, 2005 @07:02PM (#11817750) Homepage
    Why bother? If each device has it's own power brick, it's safer. If one fails, the others won't.

    If you combine all of them into a single point of failure, you might reduce what you think is an "eye sore" but at the cost of a higher risk of failure.

    Is making your setup less sturdy worth a cosmetic fix?
    • Re:Why? (Score:3, Insightful)

      by voisine (153062)
      never had a wall wort fail... ever. ac's gone out many times. I think the dc power supply is not the weak link in the chain unless you've got a battery backedup generator backed up ups at an earthquake/hurricane/tornado proof co-lo somewhere.
      • Re:Why? (Score:4, Insightful)

        by Guspaz (556486) on Tuesday March 01, 2005 @07:31PM (#11818033) Homepage
        All the more reason to continue using wall worts. You're moving from a simple wall wort to a much more complex system. One that is akin to a computer power supply. And I've had MANY computer power supplies fail.
        • Re:Why? (Score:5, Informative)

          by voisine (153062) on Tuesday March 01, 2005 @08:01PM (#11818307)
          Good point. I've had atx power supplies fail in spectacular fasion. Scared the crap out of me. I bought a used one and the local computer junk store, plugged it in, turned around, and then *KABLAM*. It was loud enough to make my ears ring. Then thick black acrid smoke started pouring out the back. I took it apart to investigate what the hell happened. It was a fairly large blown capacitor. I had no idea those could explode like that until after this little episode.
          • Re:Why? (Score:3, Funny)

            by bluephone (200451)
            Too much of the smoke escaped. Capacitors are smoke reserviors. It sounds like ther ewas a clog on the drain side of the capacitor, and the pressure of the smoke popped the capacitor. Since you say it was a used PSU, it could be that there was a build up of burnt electrons in it, and they clotted.
    • The only reason a power supply might fail is if it is built using cheap, shoddy, or underrated components. A decently made power supply will work for decades without as much as a hiccup.
      • That depends how hard it has to work, and how much power filtering it has to do. My house has dirty power. I've had half a dozen power supplies fail in the period of a year or two. One of them was a high-end Antec TruePower Gold unit, which at the time of purchase was very close to the highest-end product sold (By an already "premium" PSU company).

        On the other hand, I've never had a wall wort fail, not even the cheapest of them, made of the crapiest of components.
        • Though you do hear of the occasional recall due to potential to cause a fire.
        • by GoRK (10018)
          If you have AC that's so bad it destroys power supplies and you own your home, you should seriously consider some decent power conditioners. They cost a few hundred bucks but can seriously help your power on the whole house.
      • The only reason a power supply might fail is if it is built using cheap, shoddy, or underrated components. A decently made power supply will work for decades without as much as a hiccup.

        Are where do you think most consumers place "Wall-wart quality" only their list when they go out to buy new widgets?
        Rest assured that your wall warts (and line lumps) are a cheap as the manufacturer could get them from the third world. See the recent Dell recall for an example.

        Decent components cost money. Derating
        • by unitron (5733)
          "I wouldn't count on the cheap electrolytic capacitors in your wall warts lasting for decades."

          If it's an AC to AC wall wart, then it's probably nothing more than a step-down transformer. If it's AC to DC then they probably just added a diode or two. If you open up the piece of gear to be powered you'll probably find the filter caps (such as they are)in there (and maybe even a 7805-type regulator).

    • If you combine all of them into a single point of failure, you might reduce what you think is an "eye sore" but at the cost of a higher risk of failure.

      Actually, the reliability of each piece is multiplied together. A single product may have a 99% liklihood of surviving a year. If you rely upon 10 of those, take .99^10 and you have a 90% liklihood of surviving a year in total. Sure, a single 99% means any failure is catastrophic, but the odds are better.

      What would you plug into a wall-wart remover?

  • A non-home solution (Score:3, Informative)

    by Yobgod Ababua (68687) on Tuesday March 01, 2005 @07:07PM (#11817801)

    It's designed for large scale server rooms, and as such won't particularly adapt to what you want, but this does show that others have had the same thoughts and are applying them to various niches. They do also explain what they see as the benefits of this arrangement.

    http://www.rackable.com/products/dcpower.htm [rackable.com]

    • Also... (Score:3, Insightful)

      by Yobgod Ababua (68687)

      ...now that I think about it, so could just obtain a hobbyist DC power supply of sufficient amperage and hook all your devices up to the appropriate voltages.

      It's possible that a spare PC power supply might even suffice, but be careful that you get one that doesn't detect and auto-off when a motherboard isn't plugged in.

      Lots of power supplies from somewhere like this: http://www.kepcopower.com/prodmod.htm

  • If you do this, be sure to check for unexpected shorts. A piece of equipment we install at work has a big warning to only use it with its own power supply. Of course, it's so much easier to tie it to the beefier supply running the other panels it associates with, so many techs do just that. And it works just fine. Until...

    Turns out, due to some wierd design decision, the common pin for the serial port isn't really "common". If we plug into it with our laptops, and then plug the laptop into AC power (w
  • OMG... (Score:2, Informative)

    by malejko (216594)
    I swear reading most of these ask slashdot's are slowly killing me from the inside out.

    If you're pissed that a block takes up like 3 ports on a surge protector, get a short extension cord like this: http://www.radioshack.com/product.asp?catalog_name =CTLG&category_name=CTLG_009_001_003_000&product_i d=61-2755 [radioshack.com] , but if you want ONE DC converter with various voltages, you're just asking for trouble. An eyesore they may be, but they're quite safe for the most part and do their job properly.
  • This would be entirely possible thanks to modern power electronics, if you really want to. Honestly I'm not sure it's worth the bother, but here you go:

    Get a normal DC power supply that will supply the largest voltage you need and more current than you need. Or build one. Whatever.

    Go to some electronics shop (If you're in the UK, Maplin [maplin.co.uk] are good) and get some Voltage Regulators [maplin.co.uk]. These are basically integrated circuits that can take in a wide range of voltages and output a fixed voltage.

    Connect the voltag
    • While you are correct about the regulators being able to ourput the desired voltage, they will not usually be able to provide enough current.
      A normal LM7805 in a TO220 package can deliver 1 Amp with a heatsink (typically, check the datasheet for exact info).
      That is only 5W. In some cases (external modem) you might be able to get away with it, but in others (ethernet hub) you might not.
  • by ikeleib (125180) on Tuesday March 01, 2005 @08:49PM (#11818934) Homepage
    DC power distribution is used in telecomm applications. They use a 48V bus and use DC/DC converters to get the required voltages. The DC/DC converters are expensive, and the AC->48V converter is also expensive.

    You could just get auto-adapters for all your crap and then use an AC->12V converter. However, I imagine that this is more work than it's worth.

    Why not just do what everybody else does and get more outlet strips?
    • Why not just do what everybody else does and get more outlet strips?

      Well his original goal was cosmetic, so he wants to do something different.

      DC travels just fine over short distances. If I were him I'd just cut the ends off of the DC supplies, solder in a 10' length of wire, and put the wall-warts on the floor. Run all the lengths in a conduit up to the devices.

      Then don't step on the rocker switch on the power strip. DAMHIKT.
    • Yes, and there is a reason why they use 48V.

      I haven't seen it mentioned in the whole discussion, but a low voltage normally provides larger current.

      One of the examples above mentioned 5W, 1A. If your cable is not thick enough or too long, you will have a serious voltage drop at the end of the line, unless you can sense the voltage there and feed it back to the regulator.

      That is why 48V is used. It is not extremely dangerous (DC is more dangerous than AC because of elektrolytic effects on the blood) and t

  • I recently cleaned up my computer rats-nest at home and located no less than five wall-warts plugged in but with no equipment connected to them. Some I could remember; the old linksys router I bricked with a firmware idea, the DSL modem I replaced with cable a few months ago, and several others I just have no clue about. They join the existing crate full of cables in the garage.

    Some accessory manufacturer will catch on and start building devices that don't each have a different shaped (linksys), voltage (b
  • ICs like this [ti.com] should step down 110V to 12V, and a bridge of diodes and eeny weeny capacitor should make it DC. I just dont see why would anyone need big coils. I've increasingly been seeing 3-pin ICs that are power regulators, some that can handle plenty of wattage with a heatsink, and are TINY. I fail to understand why manufacturers are still packaging the ugly adapter. My new voip device from linksys, a PAP2, is small sleek, and comes with a giant unslightly adapter. Its the stereotypical engineer think
    • The coils are there to provide electrical isolation. If they were left off then you'd be exposed to at least one side of the 120VAC in what may be a poorly isolated unit. The coils are large because electromagnetic inductive efficiency drops with the input frequency. Efficiency is the major reason why the switching frequency in PC switching power supplies are so high, most can get away with just using air cores rather than metallic cores.

      The IC you mention is useful for completely isolated devices (no e
    • Pulse width modulators have to be fed DC so the diode bridge and some of the filtering would have to come first. However, this would still result in a power supply without transformer isolation, a thoroughly dangerous situation from both a shock and fire hazard standpoint (and using the gear with this supply instead of the one supplied by the manufacturer or at least a similar replacement may lead to a denied insurance claim if you start a fire or electrocute somebody).
    • by LordEd (840443)
      The problem is efficiency. You can drop 120V to 12V on this, but that means that the regulator has to burn up 108 V. If you're drawing 1A, then that's 108W being wasted just to step down the voltage, and 12W actually being used by the device (90% wasted).

      However, a transformer can step down voltage with a very high efficiency (google says 80 - 90% efficient).

      A stepped down signal of 12V and 1A means an input of 120V and 0.1A, plus loss due to resistance. With 90% efficiency, that means an input of abou
      • The Texas Instruments product the OP linked to is not a series pass regulator, it is a pulse width modulator, although he didn't seem to understand the proper use of it, either, as PWMs aren't designed for AC input.
        • Well, actually, the link points to a search page with one high-voltage linear regulator and two switching power supply controller chips (these aren't the all-in-one DC-DC converters, either). The linear regulator won't work due to efficiency issues and the fact that it's only rated for 150V and 120VAC would become 170VDC after rectification. A switching power supply can easily run off of AC (put in a bridge rectifier and a filter capacitor and you have DC). However, switching power supplies are complex a
    • You are smoking something. A linear regulator that converts 110V to 12V would be about 5% efficient. That means that it will run about as hot as a P4 when supplying any real load. Switching power supplies can get pretty small, but they are difficult projects even for an experienced electrical engineer.
    • It has to do with safety and cost. These are most easily met by this setup.

      Safety = human safety and equipment safety.

      If you want to run 220V (110V) into expensive equipment, you have to design for it, which makes the equipment more expensive, and should something happen, your expensive equipment can get badly (completely) damaged. Ever wondered why PC power supplies are built in sturdy inox casings ?

      Human safety is reached by providing an isolating transformer, and by stepping down the voltage to the eq

  • As far as the things we plug into wall outlets, I think it would be very wise for all electric devices that uses a transformer to use an external transformer rather than build it directly into the device. this way, if you later switch to solar you can run the device directly off the DC from the solar panel, only perhaps having to adjust the voltage. DC is preferable with solar technology, which produces DC output, whenever possible, since an invertor to convert from DC to AC cuases you to lose energy and ar
    • Please, if you don't know anything, shut up. Converting from DC to AC is just as efficient as converting from DC to DC. Unless your solar cells happen to put out the exact voltage your device requires (they don't), you still need some kind of switching power supply. Whether it takes in or puts out DC or AC is almost completely irrelevant.
      • I know I'm coming in late on this, but hey, it's an idea that's been kicking about in my head for quite some time

        You have a SLA battery bank, ok, 1.2v between each of the cells. Your charging mechanism puts out 24 volts, so you have 20 SLA cells in series.

        Follow me so far?

        You need 6v, 9v, 4v, and 12v.

        For 6v you tap off of 5 of the cells
        for 9v you tap off of . . . hrmm, 8 cells gives you 9.6, so you waste some current there or overvolt the eq.
        4 volts much the same. Except you can choose to run at 3.6 i
    • I was thinking about all the power consumed by idle wall warts around the house. The average wall wart wastes 1-3 watts or 9-27kwh a year. Not much, but multipled by the several dozen I have in operation its a lot. (Some studies have shown that "idle" electronics consume 11% of a home's average power usage)

      Most of this is becuase cheap low powered power supplies are really inefficient, often wasting more power than the device uses.

      My idea was to modify most of my devices to run on 48VDC using DC/DC con
  • Mod a computer power supply. I got my 550 Watt power supply for $10 plus shipping, and it works great. (Was supposed to be free shipping, but with that good of a deal I wasn't going to bother with it).
  • Rather than an ATX PSU that require special wiring to keep it alive, bust out your old 486 and steal the AT powersupply. It has a real power switch and enough kick to run plenty of 12v cameras and DAs (from experience). Smaller supplies can be lifted from old external HDD cases. I have one from a 20meg Mac SCSI drive that I use all the time (it is not an open frame unit).

    SD
  • It's pretty easy to build. Just need one multiple-voltage power supply that can handle the entire load. Most equipment powered by wall warts will be fine with 9-12VDC, or 24VDC. If not, you can use a bunch of adjustable voltage regulators, but each one will consume power and waste more than an equivalent wall wart. Add a bunch of cables, jacks and probably a fuse on each output and you're done.

    But is it really worth this effort just to avoid an eyesore? Just hide it behind a desk somewhere.
  • Many others have posted about the short cords that go from the wart to the plug. Another solution is a power strip like this one [apcc.com] which has a special section for wall warts where they won't cover other outlets.
  • by budgenator (254554) on Thursday March 03, 2005 @10:38AM (#11833568) Journal
    In the Army we used DC power for some applications, it's an nasty problematic way of distributing power and tends to be unreliable, if at all possible avoid it. If you are wondering why, it reinforces the fact you don't know enough about real-world power distribution to do it with out it cause more problems than it creates.
    DC is dangerous, if you get shocked by it it holds on to you more, there is no pauses in the current to allow you to get off the conductor. DC is dangerous, generaly it needs much more current for a give power because the voltage is lower, short curcuits tend to arc-weld together; Avoid DC power distribution when ever possible.
  • by Animats (122034) on Friday March 04, 2005 @01:00AM (#11841629) Homepage
    It's becoming common to use internal DC distribution within large electronic systems, then use a final DC-DC converter near the point of load. Typically, you convert the incoming AC line to 12VDC, distribute the 12VDC, then use point of load switching regulators [v-infinity.com] each place you need power.

    This, in fact, is what's happening when you have devices powered off USB, FireWire, or Ethernet. Read this discussion of how USB power distribution works. [micrel.com]

    So the parts exist to do what the original poster wants to do. But they're not typically packaged for the end user.

  • by metoc (224422) on Friday March 04, 2005 @03:52PM (#11846950)
    IEEE 802.3af aka Power-over-Ethernet and USB 2.0 both provide DC power in a standardized form factor.

    USB 2.0 is good for 5.0 volts and a max of 500ma (2.5 watts) for the bus.

    IEEE 802.3af is good for 44-57 volts and a max of 15.4 watts.

    Many devices currently on the market, such as PDAs, iPODs, and a few cell phones will charge from USB ports.

The superior man understands what is right; the inferior man understands what will sell. -- Confucius

Working...