A $5 ARM Development Board

Most of you know that there are plenty of ARM powered development boards out there, so you may not be really sure what a new one can still bring to the table.

With a $5 price tag, the open hardware McHck (pronounced McHack) is meant for quickly building projects on a small budget. The board created by [Simon] is based on a Freescale Cortex M4 microcontroller, and can be plugged directly into one’s computer. As a Direct Firmware Update (DFU) bootloader is present on the microcontroller, there is no need for external programming equipment.

The board has unpopulated footprints that allow users to add other functionalities that may be required for their future projects: a Real Time Clock (RTC), a Boost regulator for single cell battery operation, Buck and linear regulators, a Lithium Polymer (LiPo) battery charger and even an External Flash storage.

The Bill of Materials can be found on the project wiki and the McHck community will soon launch a crowdfunding campaign to send the 5th version of the board to all the hobbyists that may be interested.

And if you’re curious, you can also have a look at all the other boards that Hackaday featured these last months: the browser based IDE arm boardquad-core ARM dev board and the Matchbox ARM.

75 thoughts on “A $5 ARM Development Board

      1. Actually it’s pretty identical. It’s a microcontroller on a breakout board, with USB, a reset switch, and a status LED. Oh look it even has a boot-loader for USB loading, just like an arduino, except not one.

    1. MSP430 Launchpad has and still a lot cheaper than the Arduino boards and it even has two buttons and two usable non-status LEDs, plus it can program the whole MSP line. Gee, you shouldn’t live under a rock.

  1. “Arduino” is a rather broad term these days. If you consider that you can run the bootloader on a bare MCU, all the way down to ATtiny45 (a $0.73 at Mouser in SOIC) and just move your USB-TTL adapter from one project to the next, you might say that you can, indeed, do an Arduino project for less than $1

    A $5 ARM kit, even with unpopulated board, is till pretty cool. As things get down to the price level of a cup of coffee, the actual price loses any importance and things like community support, code examples etc. become what matters.

  2. is there plans to sell pre assembled boards?

    or boards that use the pc mount versions of the chips?

    some of us like me dont have the soldering skills or even the eyesight needed to work with parts that small.

    1. I used to think that too, but I bought a nice big magnifier, solder paste and one of those hot air desoldering stations. Apply the paste with a toothpick, put down the part with tweezers and blast with hot air, very little skill is necessary, just the proper dose of caffeine.

      1. oh and some wick braid for the inevitable solder bridges. Again not much skill, just put the wick down on the bridge, apply hot iron and wait for the sizzle. Pull away the braid and presto, the bridge is gone.

        1. I’d rather just get the thing machine-soldered at a higher quality than I will manage without having to buy any of that stuff with the parts ordered in enough quantity to where it barely costs any more anyway. If all I want to do is interface a couple sensors and actuators I can avoid SMT entirely, which is nice to my fat sausage fingers.

  3. I’ve seen boards released $8-$20 before only to jump price once users contributed code in the form of libraries..

    I hate Arduinos but it might make more sense to get one of them considering they don’t drastically inflate when their community makes their product better..

    Just going by common behavior.

    1. The base Arduino is $30, so no comparison there. These guys are truly open source, unlike some other “cheap” boards which are only cheap because they are being subsidized by the chip makers (eg TI Launchpad etc)

        1. I’d honestly rather buy products that were official arduino (support development) or obviously not official arduino (same functionality but cheaper). These things posted are trying to get as close as possible to looking official without being truly counterfeit. Feels skeevy.

          1. Nope, these are true counterfeit. They have the official arduino logo and “Made In Italy” stamps on them and everything. Pretty sure they aren’t actually made in italy though…

    2. I agree with you; this has been an unfortunate trend.

      To prevent this from happening, all project sources are under a
      copyleft license, including the hardware.

      If at any point we would turn evil (not a plan, just hypothetical),
      somebody else could produce boards at a reasonable price. In fact,
      you can already do so today!

  4. $5 is lovely, but for development you want to be able to debug. I’d like to see something like this with a versaloon built in. This is why I love my STM32F4Discovery and my ST3241G-DEVEL with their built in STLink, debugging is so simple. The open source support for STLink is excellent, I can debug at the workbench with my Linux system or in the field with my OSX laptop.

      1. I thought the whole reason for this sort of thing is to get rid of the clutter, there are many alternatives if you don’t want a bench full of gear. Might as well solder the MPU to a DIP adapter and build on a breadboard if you are going to pull out the hack wires.

  5. Where is the crystal? There isn’t one in the photo, nor on the requires parts on the bill of materials.

    This particular chip requires a crystal to implement USB. It has an on-chip PLL meant for a crystal in the 2 to 32 MHz range, and an on-chip FLL meant for a 32 kHz crystal. The datasheet warns the FLL jitter does not meet USB compliance tests. Maybe the FLL works anyway? But either way requires a crystal.

    1. Thanks Paul, I was kind of interested until I saw this. Many things “work fine” in the laboratory environment but fail in the real world with environment changes, different USB hosts, different chip batches. I would never even try to peddle something that violates the spec sheet, the people who wrote it are smarter than me, who am I to second guess them?

  6. @f another reason it fails in the real world is maybe a government conspiracy to prevent very powerful devices from being then shipped to terrorists who could then use it in something to terrorize .

    for example it is illegal to ship spark gap triggers to nuclear countries like iran because they can be used to trigger a nuclear bomb.

    and just like the power mac g4 was classified as a weapon because of it’s floating point processor and graphics card used technology similar to what the millitary uses to track rockets

    1. I don’t get it, are you saying the government actively disrupts USB communications, but only when it’s clocked without a crystal? I better go finish wrapping my computer in tin foil.

      1. That trick only works with old-fashioned tin foil made from Real Tin. Aluminum is simply an inadequate solution if you need to keep out CIA Mind Control Rays or Space Alien Radiation. Accept No Substitutes!

  7. I wish the founders good luck on the project as it has almost everything that’s needed to break into the dev board market.
    I will probably buy at least two of them but they might sit in the drawer until the board gets gdb support and can be used with Eclipse and/or has Arduino-like API.

    This is way more powerful the TI Launchpad which btw was not subsidized as some people argue. The StellarisPad might have been, though.
    TI’s boards only took off once Energia was in place, hopefully something similar will happen to McChk.

  8. It seems that you are abusing the system by ordering samples.
    Thus would it be possible that freescale won’t allow any more sample orders in the future?
    Or am I dead wrong on this?

      1. Isn’t that exactly what samples are for? To evaluate a platform? Of course you can also just buy the MCUs from a distributor or directly from freescale – they’re not much more expensive.

  9. Oh well… Only gave this a quick glance before going on with day’s schedule. Come home to see this isn’t an assembled unit. Hemiparesis still means tasks as assembling this are still a challenge.

    1. I did not find any small freescale MCUs with FPU support. That said, I think that most calculations can actually be done with fixed point arithmetic, and we plan on creating a fixed point math library.

      And we already have people working on building a quadcopter, the mchcopter: .

    10 MK20DX32VLF5 = 20$ + SHIPPING COST
    10 CKN10502CT-ND = 2.68$



    AND FOR 100 YOU PAY 500$.


    1. Maybe the post on DIY building a mchck ( http://mchck.org/blog/2013-08-06-self-built-mchck-for-5-dollars/ ) is not explicit enough on the parts cost – I apologize for the brevity. You can see that for building 5 or 10 boards (your friend wants some as well, right?), you actually have to spend less than $5 per board.

      You are right that there is no additional voltage regulator – this is because there is a 3.3V LDO integrated in the MCU itself, for use on USB or with 3 AAA batteries. You can also use a single CR2032 coin cell battery and run the board from that.

      I think you have a typo in the operating frequency: the MCU runs at 50MHz, not 40. Additionally, Paul, the creator of the magnificent Teensy3, reports that the MCU can be OCed to 96MHz.

      1. Even the ATmega328 can work at 40MHz. You will trust it?!
        Unfortunately this board will newer be out there with $5.
        The project master will newer make 10k of this boards.

        I and my friends use Arduino and Teensy for most of the development.
        Also the FRDM-KL25Z from Freescale for $12 it fare better.

        I must say that 5 bucks it’s great but unfortunately the board is just a dream to be true!
        Also the pins that are on the double row, this is a great drawback.
        The board author says that the board is for usual hacker, most of the hacks are made
        on BREADBOARDS!!! Pins in double row DON’T WORK ON BREADBOARD!

        Change that and you get a great board.
        Also the LDO insinde the microcontroller, do you man know what it’s that!?
        I guess not. Read more please before you comment.

  11. Don’t understand all he hate in the comments,

    Pretty neat idea for a $5 micro, wouldn’t be bad to buy even if it was just collecting dust. Many times I have had a project on hold because I didn’t have a spare micro laying around.

    Maybe there could be a version with an extended portion on the bottom to attach a tiny solder-less board for prototyping

  12. I don’t understand all the hate in the comments.
    The same kind of bad comments on the article talking about our MatchboxARM development board.
    Some people don’t want to understand that the diversity it’s a good thing.
    That’s why we have Arduino, Teensy, Maple, and from our team the MatchboxARM now.
    It’s like with the operating sistems, each person it’s free to choose one that best suit there needs.

    Bogdan from the MatchboxARM team.

    1. Thanks for your support! I like that your project is open source as well – this confirms my ideas and beliefs. It would be great if both projects would collaborate, especially on the software library side. Drop us a line if you’re interested!

      1. Nice to see fast replay!
        I start to work on a port for mbed online compiler to accept the MatchboxARM as a target. We can help each other on porting for your micro MK20DX32VLF5 and our STM32F103C8T6. Also we are opened also to work on library side. We think that the best approach is to have a hardware layer and abstract on top of that.
        Let’s talk more in private, enter on facebook: MatchboxARM.
        Go on the: https://github.com/NedelcuBogdanSebastian/mbed
        if you like to clone the repositories with mbed work until now.

    2. Thanks to both of you! Don’t worry too much about a bunch of hobbyists who are in over their heads and that don’t understand the differences.

      Like, people saying that arduino (at several times the cost is better), while ignoring that the typical atmega328:
      -only runs at 20Mhz vs 50Mhz here, and with lower IPC too (almost 1 MIPS/MHz vs 1.25 for this)
      -only has 2KB of SRAM which is a lot more limiting than the 8KB this one has
      -this has USB OTG compared to the *nothing* the atmega328 has (a huge deal by itself!)
      -more IOs, and with better PWM options
      -the extra Flex NVM (32KB more) and flexram (2KB extra)
      -16 bit ADCs here vs 10 bits on the 328, oh, and more channels too
      -16 bit timers (vs 8 on the 328)
      -three UARTs here vs the one on the 328
      -4 DMA channels (yes, some of us have a use for it)
      -powerful 32 bit cortex M4 with DSP instructions (and optional FPU in the MK20F) vs a weak 8 bit core

      A better toolchain, a far better set of libs all-around for just about everything, more and better bootloaders, more manufacturers to buy devices from (ARM isn’t single source unlike atmels), a far better scalability of your design (there’s FAR more powerful ARM chips!) and a better range of devices with different features and much, MUCH more!

      It’s all-around better. Sure, a Raspberry Pi would be nicer some some cases, but you can buy 8 of these for the same price!

      Thank you again

  13. I think this board is at development phase, and “build it yourself” is only meant for people that really want to play with early design (if I am not mistaken). I hope authors plan to sell this boards in kits, so people don’t have to buy 10 of them if they need only one. Soldering 0603 SMT parts is pice of cake, and can be done with any soldering iron, even without controled temperature.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.