So I have yet again over-engineered a task, this time, a freeze-dried coffee dispenser.
Why you ask, well the obvious first, it’s super fun and second, it’s the best way to learn new skills.
I’m a real noob when it comes to the inner workings of microcontrollers but I really want to know how I can use them in my projects. So far I have used the Microbits for my very personal Arc Reactor and controlling a Handheld Slide Projector.
I see some great potential in the new Microbit version 2. It has so many built-in outputs and inputs. I gave the new microphone a go this time:-)
You probably see I’m a big fan of Simone Giertz, she’s a great inspiration in doing these types of projects.
If you’re shaking with caffeine jittersand need to know if this really worked out, I can recommend scrolling down to step 6!
Stepper motor w/driver
Plexiglas for the front cover
Wood scraps for the base
A metal backdrop
3D printed parts
Step 1: Sketches…
Just a very basic idea sketched up before heading into the Java lab. I really benefit from taking the time to visualize what I want to make. And it becomes a record of my idea stream, where some ideas dry out and others like the Coffee Machine Dispenser thrives:-)
From here I know what I need to get started and I can see the first challenges.
Step 2: 3D Files
These are the files I designed to hold the coffee powder, the portioning wheel, a trigger plate for the coffee cup, and a holder for the Microbit breadboard.
I also included all parts as one STL file, which should be able to print without support.
Step 3: All Comes Together…
In this project, I got to do many of the things I love. Pretty much the first stop was the metal workshop. I knew I wanted a solid base for the parts and I bent a steel plate (went overboard already at this stage) so I would angle it to fit the cup and position the dispenser wheel.
From here I started designing the 3D parts as the central components:
- Wheel to dispense the coffee
- A frame to hold the wheel and coffee powder
- Holder for the Microbit
- And a plate to activate the switch for the microcontroller to turn on and perform its code.
During these preparations, I found a piece of leftover wood that would be a super nice plate for holding everything. I rounded the edges before I continued.
And finally wiring up and program the Microbit.
Step 4: The Cup As the Trigger
I wanted the ON-switch to be triggered by the cup being placed under the coffee dispenser. I had a small push switch I placed underneath the plate. Now I made a grove on top where the orange switch could rest.
Step 5: I’m a Programmer Now… Block Programmer!
In my hay days as a website designer I never got around to writing any code, but I got pretty good at copy and pasting code. So getting started with Microbit is a pretty smooth intro to programming a microcontroller. And let’s be clear this is block code, meaning I don’t write code I place code blocks in sequence. So don’t ask me any trick questions:-)
Don’t be fooled by the simplicity, I have friends programming Microbit drones!
The code here tells the Microbit to stay idle until a high sound is made, then it tells the stepper-motor driver to do its steps/turn, I’ve set it to do one full round, and then reset and wait for a new high sound, like a clap. (I’ve set the “screen” to show pixels according to how many steps are run, 512 for one full turn)
This is a new feature in Microbit version 2, it comes with a speaker and a microphone!!!
Step 6: Proof of Concept…
Will it work? You never know do you, for sure. This time I was really unsure. So many parts and I’m not used to this level of complexity. I also had an issue where the wheel and frame was really too tight. I tried sanding the 3D parts and some smoothness helped.
Ok so lets test this thing!!!
And what do you know it worked, IT WORKED… I T W O R K E D ! ! !
Step 7: All Good Things Must End…
After a few test runs and one nice and perfectly dispensed freeze-dried coffee things started to fail. The stepper motor had its own driver and the Microbit served as a controller, I had set what should trigger the motor and how many steps (turns) the motor would do. But what ended up happening was that the motors (4 in total) broke down. I also had a run-in with smoke coming out of the Microbit, when experimenting with increasing the power.