Hacking Dell Laptop Charger Identification

Dell Charger Spoofer

If you’ve ever had a laptop charger die, you know that they can be expensive to replace. Many laptops require you to use a ‘genuine’ charger, and refuse to boot when a knock off model is used. Genuine chargers communicate with the laptop and give information such as the power, current, and voltage ratings of the device. While this is a good safety measure, ensuring that a compatible charger is used, it also allows the manufacturers to increase the price of their chargers.

[Xuan] built a device that spoofs this identification information for Dell chargers. In the four-part series (1, 2, 3, 4), the details of reverse engineering the communications and building the spoofer are covered.

Dell uses the 1-Wire protocol to communicate with the charger, and [Xuan] sniffed the communication using a MSP430. After reading the data and verifying the CRC, it could be examined to find the fields that specify power, voltage, and current.

Next, a custom PCB was made with two Dell DC jacks and an MSP430. This passes power through the board, but uses the MSP430 to send fake data to the computer. The demo shows off a 90 W adapter pretending to run at 65 W. With this working, you could power the laptop from any supply that can meet the requirements for current and voltage.

Comments

  1. jasgio says:

    That’s why i hate laptops, only good computer is a desktop computer!

  2. Sven says:

    Nice work, i really dislike locking people into having to buy new original parts instead of repurposing old parts that fit the specifications so someone getting around the protection is always good.

    On that subject, The newer Apple laptop PSUs have the ability to start up at either ~16V 3.6A or ~18V 4.6A. There doesn’t seem to be any digital communication between the computer and the PSU. If a normal load (for example a LED driver) is attached the PSU starts up at 16V, Does anyone know what load characteristics one would need to get it to start up at 18V?

  3. adlerweb says:

    Ahhhh… Great work… I tripped over this stuff while trying to connect my laptop to a 12V-charger…

  4. valdas says:

    good one, hmmm only if it was made on atmega8 :) i have lots if them and none of msp430 :) and i have two options :D option 1. make on atmega or option 2, get some msp430 :D By the way my dell boots with non genuine charger but refuses to charge battery :)

    • Dave says:

      Having just checked my local element14, a dip packaged atmega8 is about $6 in singles. The msp430g2121 used in this is 90 cents in singles (14 pin dip).

      If you need a programmer for it, the launchpads can be picked up for about 10 bucks.

      • fartface says:

        Wow they are expensive, buy from other places atmega8 for $3.00 or less in 10 quantity on ebay as well as other sellers. You can find them as cheap as $1.95 but I think those are grey market.

      • tekkieneet says:

        I paid about $1.2 to $1.5 at QTY 10 with “free shipping” for TQFP/DIP ATmega8A/L from the usual resellers from Hong Kong. I have not have an issue with them (yet). I have seen the 328 TQFP gone down to $1.80 recently.

        If I have to buy from authorized distributors, I might as well buy the much cheaper ARM chips. The shipping charges from mail order alone can already buy me 5 of the grey market ATmega8. So guess who has my business?

  5. ejonesss says:

    or why doesnt dell just fully utilize the bms module inside the battery to include voltage and current limiting so any charger will work

    also what usually dies in the power supplies are the capacitors so they can be replaced and bring the charger back to life.

    • TacticalNinja says:

      Didn’t you just read the article? It’s all about the money, and “safety”.

      • Indyaner says:

        I agree with you that it might be majorly because of money. But also, those LED indicators are quite nice. I understand that people will choose an original psu when it got those little extras. But also, the safety-part really is a thing. I came across quite a lot of replacement PSU and am kinda stunned what people plug into their wall outlet. You get what you pay for. And if you are not willing to pay for not burning your house down, you have to suffer the consequences. Please dont play the safety-aspect down. Dell has an interest in providing a non-bursting PSU (accountability) while replacement PSU have cached their money in when you leave the store.

        • fartface says:

          The funny part, that exact charger is $19.00 to $29.00 used all over ebay and amazon.com Dell chargers are dirt cheap due to the massive amounts of corperate dumping.

          • Eric says:

            More like $8 if you have patience to wait for the container ship

          • Edward says:

            I’ve read that there are a lot of FAKE Dell AC adapters. See for yourself.

            One had several metal plates in it – not as a heatsink, but to make it heavier.

  6. yoe says:

    Nice! But the big hack we need is one in the legal system: force large manufacturers to adopt one standard charger, like what is happening with mobile phones in Europe.

    • Metalwolf says:

      unlike phones, which most people want to plug into USB ports and are generally required to meet the USB specs, laptops are like desktop computers and each have different power requirements. my old netbook uses a lot less power than my laptop. and the voltages are different as well

      • John says:

        There could be a few standardized levels. One voltage will work for all of them (somewhere between 16-20v), the only difference would be the current. They could have them in 30w, 40w, 50w, etc up to 100w. The important part would be having a standard voltage and connector, then you’d just need a charger equal to or greater than the wattage needed for your computer.

      • adlerweb says:

        …but they all meet the local AC-plug standards and use ATX-PSUs, right? ;)

    • Tony says:

      That would be this: http://en.wikipedia.org/wiki/Common_External_Power_Supply which is why phones (bar Apple) are all micro-USB now. About time too.

      There’s also one for laptops, the IEC announced a new standard a few months back.

      Actually there are two proposals for laptops, the other is to allow USB to handle 100W (typical charger is ~65W) which might be interesting.

    • adlerweb says:

      Well… Apple never adopted the standard and most manufacturers are considering new connectors since USB can’t handle enough current to charge modern phones quickly…

  7. Bogdan says:

    It may be true that OEM chargers are more expensive but it is a very wrong idea to think that the cheap replacement ones are of the same quality.
    Most of the time it is the user’s fault that the charger died and it usually has to do with overheating. People cover it with stuff on their desk or keep it covered with a blanket while using the laptop in bed, or use it on soft surfaces that don’t allow venting under it etc.

    • TacticalNinja says:

      “Spooling” the cables properly when stored is a must-know for everyone owning one. I know people who just spun the wires around their chargers (or on the cables itself), and left with a peeled cable, that looks like it has been bitten off by mice, in a few months time.

      • Paul Kastner says:

        Exactly. I crack open and solder new DC-side cables to Dell charges all the time. After reading all four pages I have an idea as to why some never recognize properly again. Thanks!

      • SavannahLion says:

        Nevermind that the design of the Dell bricks seem to imply the cables are to be wound around the brick?

        It would be difficult to find an average user who wouldn’t think otherwise.

        • phuzz says:

          All the Dell chargers I’ve seen make it very easy to wrap the mains cord end of the cable round the brick, but the end that goes to the laptop sticks straight out from the other end. The number of times I’ve seen people wrap that end as tight as they can, which generally starts breaking the cable where it exits the brick within a month or two.

      • Bogdan says:

        Yes, that one too. I know someone who’s cables(not just laptop) look like curled telephone cord.

      • metalwolfhax says:

        Im guilty of wrapping my cables around the chargers. after i damaged the covering on the cable for a few of my mobile computers, i got into the habit of zip tieing the cable on the brick with some slack on the cable. I still wrap the cables around the brick and worst case scenario, there is enough wire left over i can fix it with a wire stripper and heat shrink. I have yet to actually need to repair any cables since i started doing that.

  8. ejonesss says:

    if it is about money then why doesnt dell send out c&d notices and threaten to sue just like hollywood does with movies on p2p?

    if it is safety then why not either thermally monitor the battery on the hardware level and stop the charging if the battery gets too hot or charge it slower or switch to a safer lithium sayyyyyyy lifepo4?

    • Bogdan says:

      There are plenty of monitoring circuits in the hardware and the fact that your laptop doesn’t burst into flames even with a different power supply says quite a lot. Remember that the power brick is not just a charger, it is powering the laptop as well.
      I have a friend who worked in servicing laptops. The policy was that warranty is lost the moment when you replace the charger with a 3rd party one(some people do that although it could have been replaced by warranty) and they will not fix stuff used with third party chargers even outside of warranty.
      Cheap chargers will have worse noise, worse stability, worse efficiency, shorter life time and they will be unable to hold the rated power for long.

      • matt says:

        The power brick isnt a charger at all, laptops generally have dedicated circuitry for this. All it is, is a power supply.

      • Megol says:

        Give me one reason a properly designed computer could “burst into flames” in any reasonable realistic scenario.
        There isn’t any. Even in a cheap design the power MOSFETs and controller should be designed to tolerate a higher input voltage than the design specs – otherwise the chips are under dimensioned and will in most cases be generating extra heat.
        If the PSU can supply more power than needed? No problem.
        If it can’t supply enough power? The cheap way to handle that is just treat it as if not there, the proper way is to either power the battery changing circuits or the computer proper with a fallback to the cheap way if it still is too little.
        If it supplies too high voltage? If significantly over the design voltage input protection circuit should handle it, if not the switching DC/DC regulators should have no problem.
        If it supplies too _low_ voltage? The cheap way (as stated above) is the easiest but if it is withing a reasonable range from the design voltage the DC/DC should tolerate it too (given that the PSU then can deliver enough power at that voltage – otherwise look above).

        Doing anything else is just laziness.

  9. v00 says:

    I recently cracked open a ‘replacement’ charger after its own said that it had stopped working. I discovered that a large piece of offcut heatsink aluminum was wedged between active and neutral, hence producing a dead short on 240 Volt mains. He said he’d heard a ‘pop’. No shit! Estimated breakage current 20-30 amps. Apparently this happens somewhat frequently.

    Normally I’d be all for breaking out of manufacturer lockins, but in this case I heartily recommend people stop being so cheap and just buy the bloody OEM charger. It’s far less likely to burn your house down or kill your laptop, either of which will cost you a lot more than the cost of a decent charger. Get a second user one if you have to, just stay the hell away from no-name replacements.

  10. ejonesss says:

    unless the warranty department explicitly asks for the charger omit it without the charger they cant say void.

    probably with the sensitivity of the data on the laptop it is best to just drop another several hundred on a new laptop than to get it repaired with all the horror stories of child porn and pirated movies and software being found by the technicians.

    while i dont condone child porn or piracy there is still sensitive data that a rogue technician could steal

    • fartface says:

      Because removing the hard drive is too difficult?

      • metalwolfhax says:

        As someone who worked behind a service desk for a few months, i will say yes, it is too hard. I know people who ask me to set up their desktops even though it is generally compairable to the child toy where you put the ball in the round hole, the cube in the square hole and matching the colors. If you cant manage that, i dont want you messing around inside a desktop, let alone removing the panels on a laptop.

    • Bogdan says:

      I don’t know if it is valid for all countries, but everywhere I had to deal with warranty(never for a laptop) you are supposed to bring in all power supplies and accessories in with the product. I couldn’t get an mp3 player accepted in warranty without the original headphones which I gave away and used my own…

      • Sven says:

        That practice is illegal in many countries, it’s seen as an obvious attempt to avoid warranty claims.

        I know of people who successfully got warranty repairs from the manufacturers without any receipt, box, accessories or anything simply because the product had been for sale on the market for less than the factory warranty and thus had to be within warranty.

  11. Brandon says:

    Most Dell laptops let you turn off adapter warnings in the BIOS. Otherwise, they just produce an annoying warning.

  12. This is actually one area where that “Apple tax” pays dividends.
    For years, Mac laptops have used the same family of power supplies. And not only that, but the more powerful supplies (i.e. for the MacBook Pro) will work with the less powerful laptops (i.e. the MacBook Air). That’s a counter to the proprietary lock-in that comes when manufacturers can lock a specific power supply to a specific laptop. Owners get more choices.
    There’s actually a lively third-party market selling used Apple power supplies at about half Apple’s price. And it’s also true that your older Mac’s power supply just may work with the new one, giving you a spare.
    Used, these power supplies so cheap I’ve got three for my MacBook: one for the office, one for on-the-go, and one as a spare.
    I do have a hunch, however, that Apple’s about to make a switch to a lighter and more compact supply. The brick that ships with the feather-weight MacBook Air is ridiculous. I hope at the time they also build in USB power.
    –Michael W. Perry, author of Untangling Tolkien

    • matt says:

      I’m not sure why you’re surprised that a more powerful adapter will work with a computer which came with a smaller adapter.

      • Sven says:

        Because unlike everyone else who uses 19V or 20V (usually interchangeable) they use 16.5V for the lower power and 18.5V for the higher power. The voltage is higher for the new higher power laptops because their proprietary plug can only handle so much current.

  13. Pun says:

    Speaking of low quality power bricks, this site (http://www.lygte-info.dk/info/usbPowerSupplyTest%20UK.html) tore down and tested a bunch of non-OEM USB power adaptors. The results were interesting. Sometimes you really do get what you pay for.

  14. HackJack says:

    I know cheap knock-off power supplies from China is usually low quality. However, there are plenty of old power supplies from Dell, Lenovo, HP, Toshiba, etc. They are good quality and it just makes me sad that they don’t get a 2nd life.

  15. mike says:

    There are a number of people commenting here who do not appear to understand how smart power systems work, because communication between charger and device is an extremely useful feature. Yes, it’s annoying when the laptop limits its performance or won’t charge when the incorrect adapter is installed. But this is a safety mechanism to prevent damage to the adapter.

    A truly smart power system will poll the amount of power available from the source, then modulate the charge current fed to the battery with the inverse of the power drawn by the system itself, in order to maintain a constant power load on the charger. To ignore the available power number and charge as fast as the battery wanted would draw too much current from the adapter and possibly damage it (or at least cause it to drop out from its thermal or overcurrent protection). To ignore the available power number and charge at a safe level such that the adapter can never be damaged could add hours to the charge time and leave the full capacity of the adapter underutilized most of the time. Only by knowing the limits of the system can the charge circuitry most efficiently utilize the power it is provided.

    The battery is in no danger from incorrect adapters. There is an internal charge manager in the laptop between the DC input and the battery, and it and the battery are in constant communication so that it can provide the battery with the voltage and current that it wants (within the bounds of available power). So it’s not an issue of not using the battery’s built-in smarts, because they are indeed being used.

    Regarding alternate chemistries, for what it’s worth, LiFePO4 is nowhere near energy dense enough yet to replace lithium cobalt, unless you’re willing to settle for a third of the life of your current laptop.

    The only valid complaint here is perhaps that the specification is closed. But there are reasons for everything – I for one would not like a shady Chinese shop wrapping a 20W AC-DC supply in a box with the appropriate chip and selling it to me as a 90W adapter. The spec is closed to the general public because Dell wants to make sure the folks making adapters for their systems know their stuff, so they don’t have to fix your crap after something like this destroys it.

    No, I don’t work for Dell. But I support the decision to implement this, even if it is sometimes inconvenient. I’d rather have the system inform me of trouble with the adapter than blindly draw too much power until it eventually fries.

    • Sven says:

      The way many Lenovo laptops do it is probably the best. They have an ID system to determine if the adapter is 60, 90 or 120W, if there is no communication the computer will treat the adapter as a 60W one. This means the computer will run and charge on any 20V source with the right physical adapter, but the top power machines may be limited in maximum performance and charging may be slowed down.

    • Ralph says:

      LiFePO4 has 80% of specific energy/energy density as lithium cobalt, not 33%.

      Since you are so wrong about that, I’m not surprised the rest of your blather is mildly inaccurate as well. The laptop could easily be designed to ramp up power slowly, and if the adapter cuts out settle at a lower current value and run continuously there. No communication protocol needed!

      • tekkieneet says:

        According to A123’s website, their 18650 LiFePO4 has: Nominal voltage: 3.3V, Nominal capacity: 1.1Ah I have seen some Chinese “1.35AHr”, but not sure about actual measured results.

        http://www.candlepowerforums.com/vb/showthread.php?257543-LiIon-18650-battery-comparison

        Looking at the *measured* results, most of the tested 18650 Li-Ion have a capacity of above 2 to 2.5AHr under load.

        So far LiFePO4 has a factor of 2 less on the capacity and approximately 90% of nominal voltage. May be you can point me to *better* LiFePO4?

      • mike says:

        Beg pardon? LG et al have been making >10Whr 18650 cells for four or five years. I have yet to see a name-brand LiFePO4 18650 cell much over 4Whr. Okay, so I said 1/3 when I should have said 1/2.5… sorry? I don’t care about theoretical, laboratory-bound capacities, my figures are based on what I’ve seen to be real, commercially manufactured cells. LiFePO4 has it in the bag for power density, but in terms of energy density, it’s got a long way to go.

        And what you’ve proposed for charger detection is about the worst thing you can do, with regard to “cheap” non-OEM supplies. How long do you think a no-name Chinese power supply is going to run, on the very bleeding edge of overheat/overcurrent? Or any supply, for that matter? Nevermind that even if it survives it’ll take an awful long time to settle at an operating point, for the supply to overheat – cool – come back online – overheat – cool – etc. Your battery will be dead by then, because the damn charger will spend more time ‘off’ than ‘on’. Plus the operating point will vary with ambient temperature and amount of time the system has been on. A real nightmare from a reliability and predictability standpoint.

        Or what happens if the battery is missing or completely flat? System comes on for five minutes, you start working – oops, adapter cut out, have to wait a few minutes. I’m sure there won’t be many angry customers complaining about that.

        And as far as inaccuracies in the remainder are concerned, I encourage you to go read SBS IF’s Smart Battery System specifications and correct me if I’ve gotten anything wrong. I design things utilizing these features for a living, so if I’m doing it wrong I’d like to know…

    • NewCommentor1283 says:

      reminds me of a USB hub i have…
      the model name is “USB 2.0 Hub” but is clearly NOT a USB 2.0 device

      device listed as USB 1.1 under hardware
      XTAL on board is missing, but pads are there
      need XTAL accuracy for USB 2.0 speeds

      i doubt soldering on the XTAL would enable the 2.0
      pins are disabled when internal CLK is used anyway???

  16. Matz05 says:

    I’ve had this problem before. Old Inspiron N5040 underclocks itself and refuses to charge the battery when the (insanely fragile) DRM wire is broken or the (unprotected) lockout chip damaged, even if power quality is perfect (it was the factory charger). Modern batteries and charge circuits have safeties built-in anyway, so it’s purely a moneygrab — and one that seems to have been hurriedly added to the design as an afterthought. We ended up just buying another charger and armouring the cable rather than mess around with spoofing the challenge-response thing. Don’t buy stuff from Dell.

  17. Ed Becerra says:

    I could really use a spoofing cable like this for my Dell Mini 9 – I want to try to run it entirely from solar cells. Maybe a smart cable that’ll take any power source or even a battery.

    Can’t build one myself, but I’d be happy to talk to someone who can.

  18. Robert Glynn says:

    It’s all well and good that proprietary power supplies allow for more stringent QC, however there is one thing that hasn’t been mentioned and that is availability.

    At one point I forgot my Dell charger at home, and even in likely the most diverse retail market in the US (NYC) I couldn’t buy one. It’s great that there may have been thousands on eBay, but I needed it NOW and even at $100 it would have been worth avoiding the loss of work. I did have a PS with the exact same connector, voltage, and amperage but it didn’t work. The end result: never going to buy another Dell again (including workstations, and including at any company where I get a say).

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Follow

Get every new post delivered to your Inbox.

Join 93,718 other followers