New Part Day: Ooh, The Things You Can Do With A CLUE

There’s a new development board in town from Adafruit, and it’s called the CLUE. This tiny board can be programmed in Arduino or CircuitPython, and it is absolutely stuffed with sensors and functionality, including Bluetooth. It’s essentially a BBC Micro:bit with more sensors, a screen, and a much beefier processor. Sound interesting? Let’s get out the magnifying glass and take a look, shall we?

(Editor’s note: Adafruit ran out of the first alpha run of the hardware. While we didn’t run into any bugs, the next versions will presumably have even fewer, but will also cost $40 instead of $30. That said, they’re giving out 3,000 of them to attendants of PyCon in April, so you might also get your hands on one that way.)

And Bit:Bot takes the checkered flag! Image via Seeed Studio

First and foremost, there’s the form factor — if that bottom edge looks familiar, that’s because the CLUE is designed to work with micro:bit robot kits and anything else with that edge connector, like the CRICKIT for micro:bit, or the Bit:Bot from Seeed Studios. This is big news for the micro:bit ecosystem, and not just because the CLUE brings tons of sensors and a screen to the scene, although a 1.3″ screen at 240×240 resolution is nothing to sneeze at.

The main brain is a Nordic nRF52840, so you can pair it to your phone and stream your collected data. Or, use it to get two CLUE boards talking to each other. This is a major upgrade from the micro:bit’s nRF51822 — the CLUE is four times faster, has four times the flash memory, and has sixteen times as much RAM. We hope someone can find a way to make them into short-range messaging machines with Q10 keyboards.

If Bluetooth won’t do, you could add a LoRa node. Image via Pi Supply

Sensors Aplenty

The name CLUE is apt, because his board is all about discovery. Interested in recreating the falling eggs-periment, minus the eggs? Cradle your CLUE in bubble wrap and marshmallows and let the 9-DoF accelerometer/gyroscope/magnetometer get you all the data on its way down. This thing has a ton of environmental sensors too, like humidity, temperature, barometric pressure, and even altitude that will make it easy to prototype that weather station you’ve been meaning to build.

If you’ve been dying to get into portable game programming, the CLUE could be your board. Just think about all the different environment-based game mechanics you could use. There are also proximity, light, and gesture sensors, a speaker, and a MEMS microphone. Those two white LEDs by the USB port can do color sensing, which you could use to make a cool game for kids who are learning how to differentiate between colors.

If that isn’t enough sensors for you, there’s a STEMMA/Qwiic port that can also take Grove connector sensors with an adapter cable. Right now they’re working on a library to make it easy to access all the sensors. In the meantime, check out these examples on GitHub — more are being added all the time. We managed to snag one of these boards a few weeks ago and tried out the code examples, and it really is a nice little board.

29 thoughts on “New Part Day: Ooh, The Things You Can Do With A CLUE

  1. I do like seeing more microcontrollers come with onboard screens & buttons. It really saves a lot of time wiring and mounting those components separately – far easier for the PCB fabricator to attach them just like any other component the design calls for. That being said, I think the ESP32s with onboard display would remain my go-to: More processing power, wifi as well as BLE, and a fraction of the price depending on where you source it.

    Aside from the bigger screen, are there any other REAL advantages of this item?

    1. For getting started with a platform lots of “toys” on board may be ok but I noticed that when I have a “real project” all these over featured boards always have some sensors that come into my way by stealing pins or power.

      And sometimes even the smallest MCU of a chip family will be what you really need instead of a Deep Thought with 1337 not removable sensors.

      A naked processor and just adding what’s needed is the way that yields the best results. Maybe boards should be more LEGO-like.

      1. Something bare-bones like a teensy (bare-bones sensor wise, obviously not processor wise) or MCU breakout board plus many of the sensor breakouts or breadboard-friendly sensors (DHT11 etc.) available from Adafruit, Sparkfun, etc. seems like that’s as LEGO-like as is practical.

        But if you have the skill and the inclination to roll your own board after doing some work on dev boards (after all that is their purpose right?) then good for you.

        I guess I’d like a better explanation of what that would look like.

          1. I’ve never seen a commercial product with an Arudino in it, but I’ve personally designed plenty of fixtures with Arduino’s in them to TEST commercial products. I typically need 1-3 of each fixture for our production lines, and it’s just so much easier to throw in a $15 Arduino than roll something custom.

            I say all that to say this: It drives me ABSOLUTELY BONKERS that one of the mounting holes on the Uno is so close to the 12V port that a standoff won’t sit flush with the PCB.

            Nowhere near bonkers enough that I’d ever stop using them mind you… but absolutely bonkers none the less ;)

  2. It have A LOT of sensors, way more than most of us would use. It’s nice, but overkill for the majority of users. Unless you are doing a game where you use all those, or a blackbox for your car, it will have a lot of wasted potential.

    But it have very good things. The small LCD is a big plus. Being plug and play with a common platform is another one (and a good, but limited, use). But the price is on the heavier side.

    An ESP32 with LCD and an array of libraries to read lots of sensors would be better, cheaper, and less wasteful.

    1. This. The wide variety of I/O – display, sensors, radio – are *potential*, not wasted. You drop this into something, it means that later on when you realize you *do* want to sense gyro or temp or something else that wasn’t in your original plan, it’s available. This isn’t meant to be a processor for large production runs – it’s dev/prototyping/experimental/hobbyist/educational. Yes, in many applications 90% of the sensors will be idle, and that’s *missing the point*.

  3. Given all of the built in sensors, the choice of the micro:bit edge connector, CircuitPython, and it’s from Adafruit, I’d guess it’s aimed at the classroom educational market where a more full featured board would be attractive over something more flexible that hackers would prefer.

    But I love the nRF52840! Great little chip and so easy to use. Yeah, there are times when WiFi would be nice but not that often, for me.

    1. Agree with the educational application. After all that’s the Genesis of the micro:bit right?

      Anything that helps get people involved in technology early on and understanding how their technology works, is a winner in my book. That’s the redeeming quality of Arduino.

    1. IKR? People have been telling me that for literally *years*. Most of them know me pretty well, and understand that anything like this is way outside my budget already… and if it’s a new part, then why they knew about it for so long ahead of time really doesn’t make sense, either… especially since the Micro:Bit isn’t that old itself…

      Huh.

      Weird.

  4. As a teacher and maker, i’m happy to see this new board. The only thing that bothers me is the micro:bit connector. It’s difficult and/or expensive to work with it outside USA and EU. Why not adding a line of breadboard connectors?

  5. For USD40 you can build a board with lots of bling and extra’s, and I see why such boards seem attractive on first sight, but the more I think about such things, the less attractive they are for me.
    For USB40 you can also buy a Linux board with GHZ processor and GB of RAM.
    With WiFi or Ethernet & SSH I can also re-use the much bigger TFT that’s already on my desk.

    On development boards, you can also easily blow a few pins, which is not fun with USD40 boards.

    With USD 5 board or less, then boads get into the range of buying them just out of curiousity.
    I do not like Edge connectors for development boards, and if can be avoided also not the fine pitch stuff.
    The most universal connector is still the 0.1″ header, and IDC flat cables for connections, or direct stacking ala PC104 style.

    I also like development boards in DIP form format . They work good for breadboarding, and can also be easily put in a DIP socket on experimental boards.

    For DIP headers, I much prefer the thin round header pins over the thick square versions, and that is the reason I prefer to buy DIP form factor boards without the connectors soldered yet. Soldering them is easy.

    Also, if the boards itself are cheap, and you can easily add an TFT via a connector or flat cable, then you can simply leave the cheap boards in your projects, and add the more expensive parts only if you need them.
    Therefore I also like the 80+ sensor boxes for prototyping. You buy a box of fun stuff, and use for your project what you need.

    Sticking a small TFT on a big PCB also complicates putting the thing in some kind of enclosure.

Leave a Reply

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