Catch up on stories from the past week (and beyond) at the Slashdot story archive

 


Forgot your password?
Close
typodupeerror
×
Power Hardware Technology

Raspberry Pi Admits To Faulty USB-C Design On the Pi 4 (arstechnica.com) 134

Posted by BeauHD from the what-not-to-do dept.
An anonymous reader quotes a report from Ars Technica: The Raspberry Pi 4 was announced two weeks ago as a major new upgrade to the line of cheap single-board hobbyist computers. The Pi 4 featured a faster CPU, options for up to 4GB of RAM, and a new, modern USB-C port for power delivery. The Pi 4 was the Raspberry Pi Foundation's first ever USB-C device, and, well, they screwed it up. As detailed by Tyler Ward, the Raspberry Pi 4 has a non-compliant USB-C charging port and doesn't work with as many chargers as it should. Thanks to the open nature of Raspberry Pi (even the schematics are online!), Ward was able to discover that Raspberry Pi just didn't design its USB-C port correctly. Two "CC" pins on a USB-C port are supposed to each get their own 5.1K ohms resistor, but Raspberry Pi came up with its own circuit design that allows them to share a single resistor. This is not a compliant design and breaks compatibility with some of the more powerful USB-C chargers out there.

Whether your USB-C charger works with the Pi 4 has to do with whether it uses an "e-marked" cable. E-marked cables are fully featured USB-C cables with chips inside that negotiate power management, accessory modes, data rates, and other communication specs. Since the Pi 4 USB-C port is wired incorrectly, these smart cables will detect the Pi 4 as an "Audio Adaptor Accessory" and refuse to charge them. Usually, e-marked cables are more expensive and come with larger, higher-powered items, like a USB-C laptop. After reports started popping up on the Internet, Raspberry Pi cofounder Eben Upton admitted to TechRepublic that "A smart charger with an e-marked cable will incorrectly identify the Raspberry Pi 4 as an audio adapter accessory and refuse to provide power." Upton went on to say, "I expect this will be fixed in a future board revision, but for now users will need to apply one of the suggested workarounds. It's surprising this didn't show up in our (quite extensive) field testing program."
This discussion has been archived. No new comments can be posted.

Raspberry Pi Admits To Faulty USB-C Design On the Pi 4 More

Raspberry Pi Admits To Faulty USB-C Design On the Pi 4

Comments Filter:

  • and they saved $0.005 an unit with that.

    • How did they save any money? Besides a few demo and test units, Raspberry doesn’t manufacture the vast majority of them. Raspberry sells the designs.

      • Raspberry doesn’t manufacture the vast majority of them. Raspberry sells the designs.

        True, they don't manufacture them in the same sense that Apple doesn't manufacture iPhones. But they do not sell the design and allow anyone to sell Raspberry Pis. If you want to buy a Raspberry Pi, you either have to buy it from them, or from someone they sold finished boards to. The design is largly given away as stated in the article, most of it is open source.

    • Re:and they saved $0.005 an unit with that. (Score:4, Interesting)

      by sl3xd ( 111641 ) on Tuesday July 09, 2019 @04:11PM (#58898042) Journal

      and they saved $0.005 an unit with that.

      Were it due to an attempt at cost cutting, I might agree, but, as Eben said:

      It's surprising this didn't show up in our (quite extensive) field testing program.

      They didn't try to cut costs, they just forgot to add a $0.00156 resistor, and it seems none of the testers tried a more powerful USB-C charger.

      I imagine the field testing didn't turn up anything because the Pi Foundation gave each of the testers one of the Pi foundation's $8 chargers... because those needed testing too...

      • Re:and they saved $0.005 an unit with that. (Score:5, Interesting)

        by ebenupton ( 2424660 ) on Tuesday July 09, 2019 @05:21PM (#58898380)

        I imagine the field testing didn't turn up anything because the Pi Foundation gave each of the testers one of the Pi foundation's $8 chargers... because those needed testing too...

        Actually those turned up too late in the day, and went through a separate testing program.

        We actually grabbed a bunch of cheap(ish) chargers and cheap(ish) cables from Amazon, discarded the ones that were too terrible, and sent them out in various combinations. Because of course you worry about the vulnerability of your product to the cheap stuff everyone has, not to the expensive stuff some people have. Live and learn.

        • Sounds like USB-C is a bit of a mess, much harder to use than B. I can see how that happened, most people are going to run Pis from cheap wall warts existing phone chargers.

          • Re: (Score:3, Informative)

            by dsonen ( 1193645 )
            Cheap wall warts and existing phone chargers are what DOES work. The fault prevents "smart" chargers for high-powered devices (like newer Macbook Pros) from charging the Pi.

          • I guess they should have used a licensed Apple darkening connector (some call it lightening) instead.

          • much harder to use than B

            Actually, depends on whose perspective.

            From the users' point of view :

            - with B : you play the "hunt the correct combo of device + cable + charger that works, and also happens to charge somewhat fast". (you got an expensive high quality charger, but your device only supports 5V and the cable is shitty and can't push a lot of mA) You can also run into situation where everything will fry.
            - with C : everything will automatically negociate with everything else. (In theory, it means that your laptop could display

          • Sounds like USB-C is a bit of a mess

            No it sounds like USB-C has added complexity. But this should only surprise people who haven't read the spec. It's not messy but it does need to be understood, much like any specification that provides a wide range of capabilities.

        • Because of course you worry about the vulnerability of your product to the cheap stuff everyone has, not to the expensive stuff some people have. Live and learn.

          Which is actually one of the approach of the design of the Pi I like.
          It works with stuff that everybody has litteraly laying around. You can just grab random stuff you happen to have, plug it into a Pi and it works.

          Which brings me to one of the little disappointement I got at the announcement of Pi4 :
          both HDMI ports are of the micro type.

          I happen to be in possession of HDMI-to-microHDMI adapters, but most people I know don't have such, whereas nearly every body has some HDMI-to-HDMI cable around (e.g.: to c

        • Re: (Score:2)

          by Khyber ( 864651 )

          "Because of course you worry about the vulnerability of your product to the cheap stuff everyone has"

          Which coincidentally is cheap due to the very same fuck-up you've committed, by not following the goddamned fucking spec to the goddamned letter.

          Nice try deflecting, though. Doesn't work on those of us that know how to read and understand specifications.

      • Bullshit they didn't notice it.

        It's right there in the product announcement where alpha-testers found they couldn't plug in any random usb charger, so they built their own.

        Much easier to blame 3rd party equipment as dodgy than admit to a bug in your hardware.

      • Then perhaps charging over USB isn't a bright idea in the first place.

        I have a "good" quality USB charger at home made by a major brand, but I've found that compatibility is still pretty spotty across all devices. Some of my devices refuse to charge from anything but a USB port from a PC. Meanwhile, a standard DC adapter works every damn time no problem.

        I wish manufacturers weren't obsessed with forcing new "smart" ports down our throats, only to make multiple versions and specs anyway. They ALWAYS end u

        • No thanks. None of us pine for the days of individual adapters for each devices. Charging via USB is a great idea and having a few incompatibilities is a fuckton better than being literally incompatible with everything other than the charger it shipped with.

  • Ah, it's just lifted verbatim from Ars Technica, where the editors are intelligent enough - and proud enough of their job - to EDIT.

    As you were.

  • "cables" [...] "negotiate" (Score:4, Funny)

    by Anonymous Coward on Tuesday July 09, 2019 @03:48PM (#58897896)

    My cables dont negotiate - they fucking work or meet the cutter...

  • I mean can these people even read? It is clearly _documented_ how to do this right! The mistakes they make are an utter disgrace.
    Of course, all the fanbois will cheer them on, regardless that they have so far botched every version of this thing and needed to fix beginner's mistakes in later revisions.

    • So I take it you’ve never once in your life made a mistake when designing a computer? And how have they “botched” every version so far? Inquiring minds want to know.

      • Re: (Score:1)

        by gweihir ( 88907 )

        There are mistakes and then there are mistakes. This one is utterly idiotic and indicates a fundamental misunderstanding on how elementary electronics work. Like these people do not even know Ohm's law. Or like they are unable to comprehend the specification that clearly says how this is to be done. Pathetic.

        For their other screw-ups, they have all been obvious. If you were actually interested in them you would know them.

        • So you’re refusing to actually list any of their mistakes that you claim are “beginner's mistakes” then.

          • Re: (Score:2)

            by gweihir ( 88907 )

            You think I need to "refuse" you? Do you think you can order others around? I simply ignore your lazy and invalid defensive strategy. Read up on their gross missteps yourself. There is ample information online. Of course, there are also ample invalid and stupid excused for their incompetence, so you may have to wade though that.

            • You made an assertion, and it appears you are unwilling to backup that assertion using both evasion and deflection. So not only are you refusing, you seem to employing the tactic that people must prove your assertion wrong. That’s not how it works especially if the evidence is as abundant as you claim. Please look up “Assertion without evidence” fallacy.

              • Re: (Score:2)

                by gweihir ( 88907 )

                As the data I refer to is both obvious and easily accessible, you whole accusation is bogus.

                • So you can’t list anything that you state is “obvious” then? You can’t link to a single beiginner mistake? Seems like more evasion to me. If I research all the deficiencies that others have stated you can simply deny that’s what you’re talking about as another tactic of evasion. What you consider a beginner mistake is what’s important. Either you can state what your claim is or you cannot. Anything else is assertion without evidence.

        • This one is utterly idiotic and indicates a fundamental misunderstanding on how elementary electronics work. Like these people do not even know Ohm's law.

          The fact that you said this indicates you don't know what the problem is or why it arose. Hint: Reading the charging spec allows precisely for the configuration present here. It's not until you compare it to the spec on active cabling that you would come even close to realising why this would be a problem. I.e. You need to read a spec for something *not* part of your project.

          But you know ohms law, so you're so fucking smart.

        • Like these people do not even know Ohm's law.

          How is Ohm's law relevant? A dummy power adapter will supply the correct power to the board.

          Or like they are unable to comprehend the specification that clearly says how this is to be done. Pathetic.

          The USB-C power spec allows for three different type ports [digikey.com] : "The USB BC spec outlines three distinct types of USB port: A Standard Downstream Port (SDP); a Dedicated Charging Port (DCP); and a Charging Downstream Port (CDP) (Figure 2)." The Pi meets the DCP design. Some devices have ports that meet all three but that is not mandatory; some devices will only allow one. That's not ideal but not against spec.

  • Typical RPi stuff (Score:3)

    by ledow ( 319597 ) on Tuesday July 09, 2019 @04:14PM (#58898060) Homepage

    I have to say I always have concerns about their hardware.

    I was one of the first purchasers of the original... we discovered that the USB port dropped packets because of the shared-port design, especially if it was under heavy network load. It literally took days to discover, how it got past all the testing I can't fathom and the Broadcom engineers took forever to narrow it down (even with me and many others sending off hardware - they were convinced it was to do with the SD cards in use).

    The PoE hat had to be revised as it sometimes just didn't work. The only thing that appeared to "work" without all kinds of revisions and firmware workarounds was the DVB-T hat, and that's mostly because it's nothing more than an SPI-driven DVB chip that already had drivers in Linux for many years... there's basically nothing else on the board.

    The RPi 4 looked great. But they've seriously underestimated stock despite having a "surprise" release without any kind of hype, where they could have prepared any amount of stock in advance. You still can't get the 4Gb version at most stores, weeks after release.

    Looks like I'll have to do what I always have to do with these people... wait until the RPi 4B+++1! Rev D Turbo Hyperfighting edition of the board that's actually correct, by which time the firmware and drivers will be fully working in all the distros.

    Honestly, I'm always disappointed by their testing regimes. They sell millions of the things and yet there's always *something* on release.

    I'll wait a year or so, for the updated boards and the cases and software to catch up, when they'll be putting out an RPi5 or whatever, and then buy a handful of the working versions of the RPi4.

    I don't even get what's so hard about letting case and software makers know in advance the changes so they can have stuff there for release day. I assume it works to their advantage to "only" have the official supply and case on release day.

    • Re: (Score:1)

      by Anonymous Coward

      Yes they are simply awful. I only have 4 (pi-3/pi-2) of them in active use around the house 24x7. Awful. And that price! Horrible. Should be 4 bucks, not 40.

    • I have to say I always have concerns about their hardware.

      Not entirely surprising - its ruthlessly designed down to a very low price. The point is that "makers" don't need to worry too much about letting the magic smoke out, or ripping apart old projects to re-use the expensive computer. Most other computer makers expect you to pay premium prices for design flaws.

      Plus, its partly meant to re-capture the spirit of the Sinclair ZXs and the BBC Micro. Long waits for stock? Dodgy ports? I think they nailed it :-).

    • I have to say I always have concerns about their hardware.

      I have serious concerns about your expectations from a $30 device by a small non-profit foundation. Get a grip man, this isn't a high performance product from a multinational corporation leveraging vertically integrated supply chains.

      I have never had concerns about their hardware or its supply. I am continuously amazed at what the Raspberry Pi foundation has and is achieving.

  • So, naturally, I wonder if there's an easy fix to this...

    Maybe we can mod the Pi and include the missing resistor (this assumes that the solution is indeed this easy) ?
    Or possibly just use a pass-through dongle inbetween the Pi and the USB-C charger that corrects the identification mistake?

    I'm keen to see what the hacker hive come up with, since this is a product that is (at least half) aimed for these users.

  • Schematics are not fully available (Score:3)

    by dskoll ( 99328 ) on Tuesday July 09, 2019 @08:57PM (#58899302) Homepage

    The Pi schematics are incomplete and heavily-redacted. It just so happens that the available schematics included the USB-C port and associated circuitry, but the Pi hardware is not open by any means.

    That said, the Pis are sweet little machines. I own a Zero E, a couple of 3Bs and a 4.

  • They really should have used a barrel connector.

    So many naive Raspberry Pi owners wonder why their fancy gadget randomly corrupts SDCards and get the flashy undervoltage icon thing in the top-right-hand corner of their screens when using their phone charger.

    Orange Pi had the right idea with the barrel connector. It's ponderous why the Raspberry Pi Foundation didn't address this years ago, especially with the market flooded with sketchy USB Type C chargers [gizmodo.com].

    • Re: (Score:2)

      by AmiMoJo ( 196126 )

      Probably because it's easier for people to blow their Pi up with a barrel connector. Is it centre positive or centre negative? 5V, 7.5V, 9V, 12V, 15V? Adding a wider input range supply would have added cost.

      Also, people have USB chargers so it further reduces the cost to the consumer.

    • They did address it. The singular requirement was for the Pi to use a powersupply that users likely already had further driving down cost.

      If you want barrel connector, buy one and connect it to the GPIO header. You want it battery powered, connect it to the GPIO header, you want PoE > GPIO header.

      Raspberry Pi gives you any option you could possibly desire for powering their device, but by default it ships with the one of use to the most people.

    • The main reason I can see for not using a barrel connector is that there are many different ones and those are the standard ones. With USB-C or micro-USB, there's only one shape and size.

  • My guess is that people were like "oh, this cable doesn't work, let's get another one, oh, this one works, guess that other cable has a problem even though it works with my phone, can't possibly be a problem with my Pi"

Slashdot Top Deals

You know you've landed gear-up when it takes full power to taxi.

Close