Kuro5hin.org: technology and culture, from the trenches
create account | help/FAQ | contact | links | search | IRC | site news
[ Everything | Diaries | Technology | Science | Culture | Politics | Media | News | Internet | Op-Ed | Fiction | Meta | MLP ]
We need your support: buy an ad | premium membership

Wanted: low power DC standard bus

By Paul Johnson in Op-Ed
Mon Mar 12, 2001 at 06:55:22 PM EST
Tags: Hardware (all tags)

I've been purchasing a lot of portable electronic equipment for work. At present I have an Ethernet hub, a USB hub, a digital camera, a laptop, and a home gateway (small box containing mini-firewall). But power supply is a headache.

All these devices come with an external plug-cum-transformer or some similar power supply which often weighs more than the device its supposed to power. They all produce different voltages and require a UK power socket. I need to travel abroad with this collection but organising the correct power is going to be a headache. They are all wired or built in to standard UK plugs, and most won't work off 110 volts. At the very least I'm going to need a step-up transformer and UK power strip.

Oh, I almsot forgot, the one part of these things that they have managed to standardise is the plug that conects the transformer to the gizmo. So plugging a 24V PSU into a device that expects 6V is a horribly easy thing to do.

What I need is devices which have a common standard low voltage power supply, so that I can have one single transformer and plug all my devices into that. I envisage something that could produce maybe 10 amps DC at each of 5, 12 and 24 volts from any AC source in the 110-240 volt range. That way I would only have to take my single standard PSU, or better yet I could ask my hosts to lend me their PSU because this would be a world standard and everyone would have one.

So, does anyone know of any kind of forum where this sort of thing can be raised or discussed? Does any one work for companies like DLink, 3Com, Psion, Canon or Dell (the manufacturers of my particular kit) who could start the ball rolling on this? Or am I going to be stuck with paying for separate transformer plugs with each piece of kit I buy and then lugging them around the world?


Voxel dot net
o Managed Hosting
o VoxCAST Content Delivery
o Raw Infrastructure


What about a standard low voltage power system?
o Lets start an open source project to design it. 11%
o I'll table it at the next ISO/ANSI XYZZY committtee meeting. 6%
o Its irrelevant: the market doesn't want it. 6%
o Its a good idea, but it will never happen. 41%
o The low voltage subgroup of the Tripartite Illuminati will conspire to stop it. 29%
o I work in one of those companies and I'll pass this idea up the chain. 0%
o The standard gets ratified next week you ignorant idiot! 3%

Votes: 77
Results | Other Polls

Related Links
o Also by Paul Johnson

Display: Sort:
Wanted: low power DC standard bus | 22 comments (20 topical, 2 editorial, 0 hidden)
What I do (2.50 / 2) (#2)
by Da Unicorn on Mon Mar 12, 2001 at 07:27:42 AM EST

I'm not from the UK nor do I travel there so this may not work.

When I travel witth portables I take a standard power strip surge suppressor combination. I plug all the wall warts into that strip gizmo and it goes in my carryall first. The plugs are color coded with color dots on the device although my plugs are not all the same. For international travel you would need an adaptor for the power strip plug I suppose.

I would be more concerned with improper polarity than improper voltage. Some supplies put out AC some DC so thats another issue. Most devices can handle an over voltage condition and some even have built in regulators to ensure a steady voltage.

You are not alone in this conundrum. Having a wonderfull portable device that has a power supply that needs its own transportation certainly sux.

Your mileage may vary.


Do it yourself? (3.00 / 1) (#3)
by h2odragon on Mon Mar 12, 2001 at 08:32:39 AM EST

Most of your wall warts are likely 5vdc and 12vdc; add up amperages, get connectors, and build a BMF power supply.

The problem (for the manufacturers, too) lies in the oddballs that need 16 volts, or AC, or even +4.5vdc / -9vdc (what one of my wall warts on the handy power strip says). Those devices get designed expecting a weird power supply because it's cheaper to commision a custom run of wall warts than it is to use standard DC to DC components, I gather.

Plugs (3.00 / 2) (#5)
by wiredog on Mon Mar 12, 2001 at 10:44:28 AM EST

Man, people have been complaining about this for years. Another complaint is having the transformer right on the plug instead of in the middle of the cable, where it won't interfere with other plugs.

The idea of a global village is wrong, it's more like a gazillion pub bars.

solution (5.00 / 1) (#8)
by h2odragon on Mon Mar 12, 2001 at 02:01:14 PM EST

There's a 1 foot extension cord available to solve that problem; I buy mine from cyberguys.com (no link because they require Java for store navigation, somebody with an alternate source who doesn't, please post).

[ Parent ]
DIY electronics (2.00 / 4) (#6)
by Signal 11 on Mon Mar 12, 2001 at 12:18:31 PM EST

Something like this wouldn't be too hard to do yourself.

I'd recommend winding yourself an autotransformer and tapping it at the voltages you want. An autotransformer is a coil (often a toroid) with taps at various points. It is called an autotransformer because it is *one* coil, and not two. That means it can't be used to isolate the circuit. Bear this in mind!

Let's say you make a coil with 60 turns, and you need it to supply about 2 amps. Use 22 gauge wire to wind the coil. FYI - good engineering practice would have you go to 20ga wire because the max for 22ga is 2.1 amps. :)

As you wind it, tap it out at the appropriate voltage. Let's say with 60 windings you want voltage outs of 12, 9, 5, and 1.5 volts. You want taps then at the 5th, 7th, 12th and 40th windings, respectively. The voltage drop per winding is equal across the coil, hence the numbers here.

After that, put a full-wave rectifier on each output and tie it to ground. A 1000uF capacitor in parallel with the OUTPUT side of the rectifier should be sufficient to smooth the output. If you want better filtering, there are a variety of filtering solutions out there - LC filtering in particular is good for high loads. You should now have (pretty) decent DC outputs at each voltage. If you need better regulation, use a voltage regulator - most any will do up to about 40V output, although a heatsink may be necessary.

Alternatively, if you don't need to have voltages at every output active simultaniously, I'd just pick up a 12V transformer somewhere (or wind it yourself), stick a 3A full-wave rectifier on it and a voltage regulator with the feedback tied to a resistor network and a switch.

Either way, the cost of this wouldn't exceed about $10. If you want a more professional solution, use an old PC AT power supply.

Society needs therapy. It's having
trouble accepting itself.

Whups! (5.00 / 1) (#7)
by Signal 11 on Mon Mar 12, 2001 at 12:21:55 PM EST

I should note an assumption I made - I was assuming 120VAC / US voltage. If you're in the UK, you'll need to recalculate where you tap it. If you need something that's compatible in multiple countries, you can't go wrong with IC voltage regulators, although you may need a few of them in parallel to cope with the load....


Society needs therapy. It's having
trouble accepting itself.
[ Parent ]

You have given just enough info to be dangerous.. (4.50 / 4) (#11)
by Zukov on Tue Mar 13, 2001 at 08:52:55 AM EST

If you are going to recommend people wind their own power transformers, you should try to point out at least some of the many hazards of doing so. The right choice for core material, wire gauge, and number of turns is a matter of extensive calculation, not trial and error. The choice of insulating components and details of construction is a matter of experience, which the poster of the question does not have.

The choice the transformer core (material and saturation flux) is vital. You probably realize that transformers for 50/60 Hz use iron plates for the core. Someone who picks a small ferrite toroid and winds wire around it, and then plugs the result into the wall is going to

  • burn themselves when the wire catches on fire
  • electrocute themselves when the insulation on the wire burns off
  • blow the circuit breaker.

    Even if the first attempt is not a spectacular failure, the inner windings of the transformer can get hot enough to eventually break down the insulation, causing fire and electrocution. Testing a prototype with thermocouples wound into the transformer during construction is again outside of the ability of the person who posed the question.

    Getting DC from the taps needs only one bridge on one tap. The output from any other tap to the output of the bridge will be DC (rectified AC).

    Finally, an autotransformer, which as you rightly point out is non-isolating from the mains voltage, is not safe in this application because the low voltage devices depend on being isolated from each other as well as from the mains voltage for the safety of the user as well as the safety of the devices to each other and to any outside connections such as phone lines and network lines.

    A proper transformer for this applicaion would have an primary input winding for the mains power, and completely seperate secondary windings for each of the devices to be powered. Even devices which use the same voltage should not share the same output winding, they should each have their own winding so as to be isolated from each other power-supply wise.

    Then again, maybe your post was a deliberate troll to see who would rise to the bait, although I would hope you would pick more obvious subjects, such as

    "Give yourself an appendectomy with only a stanley knife and a bottle of jack daniels"

    ȶ H (^

    Yes, I have just bumbled upon Gnome Character Map. Please ! me.
    [ Parent ]

  • Twit. (1.00 / 3) (#14)
    by Signal 11 on Tue Mar 13, 2001 at 08:32:58 PM EST

    Someone who picks a small ferrite toroid and winds wire around it, and then plugs the result into the wall is going to...

    design computer power supplies for a living, you moron. Open one up some time and trace the fucking leads.

    Society needs therapy. It's having
    trouble accepting itself.
    [ Parent ]

    Ummmm...Kay... (none / 0) (#17)
    by Zukov on Wed Mar 14, 2001 at 12:39:43 PM EST

    Thank you for your well reasoned, informative and articulate reply.

    As you may be aware, computer power supply _transformers_ do NOT operate at 50/60 Hz. The AC input from the mains is rectified to make DC at a high voltage, often 300V. This high voltage is then switched at Kilo-Hertz (I'm spelling this out just to make sure you understand) frequencies, and this high frequency AC is well suited to transformers with toroid ferrite cores and relatively low winding counts. This is where the term "switching power supply" comes from.

    Unfortunately, the person who asked the question wants a transformer which works at 50/60 Hz directly from the mains. Ferrite toroids are not suitable for use at this frequency. In addition, the lack of isolation from the mains makes your suggestion a death trap.

    HTH. HAND.

    ȶ H (^

    Yes, I have just bumbled upon Gnome Character Map. Please ! me.
    [ Parent ]

    Well... (none / 0) (#19)
    by Signal 11 on Wed Mar 14, 2001 at 01:42:46 PM EST

    Ferrite toroids are not suitable for use at this frequency.

    You could have simply pointed out I had oversimplified the circuit (which is true, now that I look over my original post) and said "you need to add these pieces..." to make it work. Had you done that, you would have saved us both a considerable amount of frustration. A toroid may not be the best approach, and since I *DID* say this was DIY (read: amateurish) it's assumed you're going to need to tweak things to get it to work. I'm assuming after you cleared the smoking remains of the previous coil of your breadboard you would be intelligent enough to try something else. Lastly, the increase in frequency probably has more to do with making the capacitors more efficient at filtering - higher frequencies are easier to filter and you can use smaller capacitors. This (the higher frequencies) isn't necessary for power supplies that are only supplying a light load that doesn't fluxuate much. I know because I have one sitting four feet away from my computer sitting on my workbench and it works fine for my little logic circuit run off of a bunch of op-amps. :^)

    The isolation from the mains is somewhat of a moot point because a sensible person would have the power supply shielded so you can't stick your fingers into it, etc. I'm assuming this is why you're upset about the lack of isolation - so that they need to be touching both outputs instead of just one and a ground to receive a shock.

    Society needs therapy. It's having
    trouble accepting itself.
    [ Parent ]

    Pls see question at end of my post. (none / 0) (#20)
    by Zukov on Wed Mar 14, 2001 at 03:20:47 PM EST

    Sure, the higher frequency is easier to filter with capacitors, but higher frequencies are able to work with small transformers and this is key to the design of switching power supplies.

    Look at the inductors (little coils) inside radio equipment, (these are used at MHz+ frequencies) you will see many of them have no core inside the coil, they are "air core" inductors. Power supplying circuits operating at progressively lower frequencies need cores with progressively larger inductance and saturation flux to control the idle current so they use ferrite cores at KHz freq. (to control core eddy current losses, which are high in iron cores at KHz freq.) and then iron plate cores at 50/60 Hz. The iron is in insulated plates again to control eddy current losses.

    Look up the equations for AC current flow thru inductors and transformers and you will see that they are frequency specific.

    To give an analogy that can be visualized, think of

  • an inductor as a mass
  • A capacitor as a spring
  • resistor as a friction device.

    In this analogy, the force on a mass is the voltage applied across an inductor. The movement of the mass represents the the current flow thru the inductor.

    If you want to try a bench-top experiment, use a short-circuit protected sine wave generator (not a square wave generator) to drive a inductor at a constant current (or voltage), and measure the voltage across (or current thru) the inductor as your change the frequency over several orders of magnitude.

    Finally, the isolation danger is not because someone might stick their finger in the power supply. It exists because the "low" voltage DC which is supplied to the load may be hundreds of volts with respect to ground, or to another "non-isolated" device which is being powered from a different leg of a multi-phase supply. The primary to secondary "isolation" of a transformer removes this hazard.

    I know because I have one sitting four feet away from my computer sitting on my workbench and it works fine for my little logic circuit run off of a bunch of op-amps. :^)

    If you are saying you have 60 turns of wire around a random toroid you had lying around and have plugged this into 120V AC at 60 Hz with nothing in series to limit the current (like a light bulb) then I would say you have been very lucky in your choice of toroid. What idle current does this setup draw? I would suggest you get some sort of AC wall-wart transformer to replace your setup, to avoid the serious danger from not having the circuit isolated.

    ȶ H (^

    Yes, I have just bumbled upon Gnome Character Map. Please ! me.
    [ Parent ]

  • toroids.... (none / 0) (#21)
    by Signal 11 on Wed Mar 14, 2001 at 07:18:55 PM EST

    then I would say you have been very lucky in your choice of toroid. What idle current does this setup draw?

    No more than 5 watts, because that's the maximum output from the last opamp. It's a 12V supply... :) I should know this, but I'm not sure how much current it draws without any of the inputs "live". I would guess ~200mA. Like I said, I'm an amateur and I have my "bucket 'o' parts" in the corner to prove it. They're all charred and blackened, by the way. heh. As far as the toroid, it has a huge cross-section because I made it myself. So it's like 10x the size of a "normal" toroid and it's wound on a chunk of some random metal I found and heated up to get it approximately roundish.

    Society needs therapy. It's having
    trouble accepting itself.
    [ Parent ]

    You made an iron (steel) core toroid.. (none / 0) (#22)
    by Zukov on Thu Mar 15, 2001 at 11:16:23 AM EST

    I take it that ends of the bar do not quite touch, or have some insulating oxide on them, or something else to avoid making a shorted turn.

    Iron torids are sometimes made out of iron tape, sized (approximately) like the tape out of a tape measure.

    Wind many turns of the tape around a mandrel (form), insulate the layers from each other.

    As you are into electronics, you might like "Nuts and Volts" Magazine. There are a lot easy circuits in it, and many of the surplus electronic places have ads in it. There are also articles about melting and casting aluminium.


    This is the book I started with (but I have the first edition) to understand electronics. You will save lots of money on charred parts. I'm not suggesting you get it from Amazon, they just make a conveninent pointer.


    Finally, no inventor is complete without a book from Lindsey..


    ȶ H (^

    Yes, I have just bumbled upon Gnome Character Map. Please ! me.
    [ Parent ]

    Power supply isolation (5.00 / 1) (#15)
    by Paul Johnson on Wed Mar 14, 2001 at 05:44:45 AM EST

    the low voltage devices depend on being isolated from each other as well as from the mains voltage

    Hmm. I hadn't thought of that. When I talked about a DC standard bus I envisaged a module with its own PSU and then something like a standard AC power strip in miniature so that several devices could draw power from a common bus.

    I can see that currently there may be devices which tie "Signal ground" to something other than their internal 0 volt line, leading to an interesting situation when you connect them together. But presumably this could be overcome if the designers were working on the assumption of a standard power bus.

    You are lost in a twisty maze of little standards, all different.
    [ Parent ]

    Not just signal ground issues.. (5.00 / 1) (#18)
    by Zukov on Wed Mar 14, 2001 at 12:49:04 PM EST

    Not just that, but internal clearance and creepage distances between items on the circuit board have been designed for the low voltage, since the isolation from high voltage is done at the plug in power supply.

    Ditto for any switches, touch pads, displays, etc. Some devices (like modems) may have an isolating transformer on the outputs, but the isolation voltage rating of that transformer has been chosen according to what's needed on the phone line, not to provide protection from the mains voltage.

    ȶ H (^

    Yes, I have just bumbled upon Gnome Character Map. Please ! me.
    [ Parent ]

    Firewire? (3.00 / 2) (#9)
    by delmoi on Mon Mar 12, 2001 at 06:59:06 PM EST

    Unless I'm mistaken, firewire carries it's own electrical signal on the wires with the data, so all your external periprerals don't need their own cables/batties. I'm not sure about laptop firewire though, there are two kinds of cables.
    "'argumentation' is not a word, idiot." -- thelizman
    Safety (3.00 / 2) (#10)
    by ShawnD on Tue Mar 13, 2001 at 08:00:12 AM EST

    Remember that one reason that 'Wall Warts' are so popular is they make UL, CSA and other approvals much easier. A low power devicce powered from an approved supply likely needs no approvals.

    If you had a fire due to a home-made power supply I am sure your insurance company would have some questions. Also the manufacturer would probably consider the warrenty void if you don't use their power supply.

    DC generation: Standardize voltage (4.00 / 1) (#12)
    by bmasel on Tue Mar 13, 2001 at 02:53:09 PM EST

    Given that electronic devices actually run on DC, is there a legitimate reason they are engineered for so many different voltages? Standardizing at 12 volts would make it all operable on automotive power supplies, and avoid inefficient inverters for those who live off the grid.

    I am not currently Licensed to Practice in this State.
    12V not enough (5.00 / 2) (#13)
    by jkominek on Tue Mar 13, 2001 at 03:42:47 PM EST

    car power is noisy noisy noisy! the only thing that the 12V your car produces is good for is running through a voltage regulator to get some voltage below 10V or so. (9V and 5V)

    trace engineering produces 94% efficent inverters, btw.

    a standard DC power bus would have to provide a couple of standard voltages, or one fairly high voltage that everything could safely cut down with a $1 voltage regulator. (I think 24V would probably suffice)

    - jay kominek unix is all about covering up the fact that you can't type.
    [ Parent ]
    Tolerance must be defined. 12 volts "car" (5.00 / 2) (#16)
    by NKJensen on Wed Mar 14, 2001 at 07:22:36 AM EST

    The question is about :

    HOW to get a standard plug, a standard voltage tolerance and customer acceptance of such a new low voltage bus.

    I'm an Ms EE myself, and this question is not a tech issue. It is all about politics. There are no serious physical problems to be solved here.

    Some big shots in Philips, Sony and IBM could just take a stand in this matter and get it to market with or without an IEEE standard to go with it.

    The rest of the world would follow shortly (except the French) thereafter, eliminating the need for adapters / Wall Warts.

    As for the 12 volt "as in cars", no can do, son!

    The "12" volts in the car can be anything from 12 to 15 volts under normal driving reaching extremes below 0 volts and above 40 volts during starting or electical failures between battery and generator.

    A minimum requirement to a voltage supply is the tolerance, the impedance and what to do with overloads.

    Modern switching power supplies are 95% efficient or better and they can step up or step down the voltage as needed for the application. What I'm saying is that the voltage does not matter as long as it is below the limit of safe handling by humans - i.e. about 40 volts DC.

    So please, big company guys, have a chat over a cup of beer some day and just define the low power DC bus the world wants so badly.

    From Denmark. I like it, I live there. France is another great place.
    [ Parent ]
    Wanted: low power DC standard bus | 22 comments (20 topical, 2 editorial, 0 hidden)
    Display: Sort:


    All trademarks and copyrights on this page are owned by their respective companies. The Rest 2000 - Present Kuro5hin.org Inc.
    See our legalese page for copyright policies. Please also read our Privacy Policy.
    Kuro5hin.org is powered by Free Software, including Apache, Perl, and Linux, The Scoop Engine that runs this site is freely available, under the terms of the GPL.
    Need some help? Email help@kuro5hin.org.
    My heart's the long stairs.

    Powered by Scoop create account | help/FAQ | mission | links | search | IRC | YOU choose the stories!