From Zero to Nano

Have you ever wanted to build your own Arduino from scratch? [Pratik Makwana] shares the entire process of designing, building and flashing an Arduino Nano clone. This is not an entry-level project and requires some knowledge of soldering to succeed with such small components, but it is highly rewarding to make. Although it’s a cheap build, it’s probably cheaper to just buy a Nano. That’s not the point.

The goal here and the interesting part of the project is that you can follow the entire process of making the board. You can use the knowledge to design your own board, your own variant or even a completely different project.

from-zero-to-nano-thumb[Pratik Makwana] starts by showing how to design the circuit schematic diagram in an EDA tool (Eagle) and the corresponding PCB layout design. He then uses the toner transfer method and a laminator to imprint the circuit into the copper board for later etching and drilling. The challenging soldering process is not detailed, if you need some help soldering SMD sized components we covered some different processes before, from a toaster oven to a drag soldering process with Kapton tape.

Last but not least, the bootloader firmware. This was done using an Arduino UNO working as master and the newly created the Arduino Nano clone as target. After that you’re set to go. To run an actual sketch, just use your standard USB to UART converter to burn it and proceed as usual.

Voilá, from zero to Nano:

If you still want to build an Arduino but the difficulty level is a bit high for you, maybe a good idea is to start with the Shrimp.

20 thoughts on “From Zero to Nano

  1. >ferric chloride
    >Year of Our Lord, MMXVII

    Come on son, where the copper chloride at?

    Anyway…I really need to start using the autorouter more and stop being so anal about having “perfect” routing. Random, unconnected bits of copper pour aside, “press button, receive layout” may be worth overlooking the oddly-run track here and there.

      1. The thing is, unconnected copper can act as an antenna, causing undesired operation of your design. More of a problem for RF and high-speed circuits, I’ll admit, but still.

        And “never using up etchant” is, handily, the main draw of cupric chloride: Using it causes it to make more of itself. I never worry about how much etchant I use, because I just dump all of it back into the jug when I’m done.

      1. despite from being ugly, autorouters can also be evil (the circuit not performing the way you intended) but this does not need to be so if you use the autorouter wisely. Here it seems to be click-and-forget, wham-bam-thank you autorouter. But despite from all that when it works for the one using it, well there isn’t any problem, or will these problems come later…

        Personally I see drawing a PCB as a work of puzzle (I always want it to be a piece of art… but it never comes out that way. Making a PCB neat and tidy, keeping important traces short or thick while other traces can be longer or thinner, but that’s a different discussion. Anyway, drawing the PCB can be as important as drawing the schematic. I’m sure that it’ll work, but next time keep the traces of the Xtal with it’s loading capacitors short, keep your ground as a whole, place decoupling close to the parts that require it. Then you will notice that strange behavior (should you ever encounter it) may magically disappear.

        Anyway, it is a nice demonstration of what you can do at home with the right stuff and that’s the proper spirit, so thanks for sharing that.

    1. Back when I used Eagle a lot, I hot-keyed autoroute and ripup. I’d alternate between the two, fixing up the worst-looking traces as I went along, and providing it with hints when it got stuck. This kind of man-machine cooperation actually worked great. Start there?

      I’ve used Kicad for four years now. It has automatic enforcement of design rules which makes the above basically pointless. And now that push-and-shove has been workable for a couple years, I can’t imagine going back. (That said, I’m sure it would work out just fine.)

      I haven’t seen an autorouter that wouldn’t benefit from human intervention, even in terms of just the logic of the layout. Aesthetically, it’s a hands-down for the fleshbots.

    1. This isn’t even close though. Did they mine the raw materials to make the raw silicon themselves?

      Reminds me of the guy who spent $1500 to make a sandwich literally from scratch.

      1. It is not the same with the arduino. If you buy the real one, it is about 10X more expensive than buying the atmega and whipping out a PCB. But if you can wait, yeah you can get a cheap one from China. Neither version will give anything back to the creators of arduino in terms of money, but they help drive the sales up.

        Pff, I still remember the times before arduino when the first things people were building when starting with micros was a simple dev board.

  2. I’m still learning KiCad and am disappointed in the lack of a simple to use autorouter. I have read lots of comments about how bad they are and fully intend to become proficient at routing without the autorouter. I’m coming from using perfboard running the leads through the board and then using them as the wiring, adding jumpers where necessary. The autorouter has got to beat that right?!?! It would be nice to have as a stepping stone with every intention of graduating away from using it.

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