Introduction: Alexa Controlled Coffee Maker
This is a work in progress. Not all of the functionality that I would like to have is part of this yet but Alexa will at least turn on my coffee machine and brew a cup of coffeeAlexa.
A few assumptions before we begin. This tutorial assumes you have used a Raspberry Pi and Arduino before, and that you are familiar with using that command line. You dont need to know much about coding in Node.js or Python, the two languages used as the code is written already but some understanding may be useful.
1. Raspberry Pi 3
2. Arduino Uno
3. Soldering iron and solder OR hot glue gun
5. 5 wires
6. Farberware Coffee Maker (Really any simple coffee maker that has a push button start will likely work)
When everything is set up, the flow of instructions from user to hot coffee goes like this:
1. User tells Alexa "Tell Coffee Maker Brew"
2. Alexa relays the command to the CoffeeMaker skill (which is a lambda function) in Amazon Web Service (AWS)
3. The CoffeeMaker skill runs its code, in our case it sends a POST message to a public url that looks like: `http://XXXXXXXX.ngrok.io`. Where XXXXXXXX is some hex value. The exact url is given after you run the ngrok command in command line (more on that in a bit).
4. ngrok then tunnels to the localhost of the machine that started it (in our case a Raspberry Pi). By default it tunnels to port 5000 but for our demo here its mapped to port 5002 of the pi.
5. The Pi is running a Python webserver on localhost:5002. When it gets a `POST` on `/coffee` it send a command to the Arduino that is connected via USB using nanpy. nanpy allows you to send commands to an Arduino using Python, which is handy since our webserver is written in Python.
6. The Arduino then sends a signal to a relay that is connected to the `On` button of the coffee maker.
7. About a minute later, hot coffee is pouring into our cup.
Step 1: Software Setup
**Lets Get Started**
All code for this project can be found on Github. Im not going to write how to set up AWS or Alexa. There are some excellent tutorials (which I used) that do a way better and more thorough job then I ever could.
Setup instructions for nanpy can be found here.
AWS and Alexa Skills setup (along with code I ended up modifying for this) can be found here. For this project you can skip any part about using `Particle` mentioned in that tutorial. I suggest setting up AWS and the Alexa Skill before continuing on.
One note if you use the tutorial above for setting up the Alexa Skill. The tutorial uses Particle to turn things on and off, so the commands that are set up for the Alexa Skill are `on` and `off`. We dont want to use Particle, we want to control our coffee maker. The current state of this tutorial is such that it is not programmed for `off` though it could be easily adapted to add this functionality. Also, I used the keyword `brew` for my command. Towards the end of the Particle tutorial in the Setup Skill in the Develop Portal in steps 6 and 7, replace the word `Particle` with `Coffee Maker` and in the Interaction Model section, step 4, use the text `CoffeeMakerIntent brew` instead of the values given.
Using the Raspberry Pi navigate to https://ngrok.com/. Click the download button and select the 32-bit platforms option and select the Linux 32-bit download (this tutorial assumes you are running Raspian on your Pi, Raspian is a Linux variant).
Unzip the file using `unzip /path/to/ngrok.zip` command. Next navigate back to the ngrok home page, click on the `Sign up` button. You should see a few steps. You have already done step 1 (download ngrok). Copy the command in step 2. It should look like `./ngrok authtoken <some_token>`.
Navigate to the folder that you saved ngrok to when you downloaded it. Paste the command including your unique token into the terminal and hit enter. Finally run the command `./ngrok http 5002`. 5002 is the port that our webserver will be listening to. You should see something like:
Session Status online
Account (Plan: Free) Version 2.2.4 Region United States (us) Web Interface http://127.0.0.1:4040 Forwarding http://562f96bb.ngrok.io -> localhost:5002
Connections ttl opn rt1 rt5 p50 p90 0 0 0.00 0.00 0.00 0.00
Copy the hex right before `.ngrok` on the line that says `Forwarding`. In our case it's `562f96bb`. You need to modify a line in `AlexaCoffeeMaker/ASK/src/index.js` from the github project. Line 86 of the file is `hostname: 'def5fe6d.ngrok.io',`. Replace the hex in front of `.ngrok` with what you copied from running ngrok. In our case the new line looks like `hostname: '562f96bb.ngrok.io',`
Save the file, zip it with AlexaSkill.js and upload it to the Alexa Skill in AWS and save it in amazon.
Here are the instructions for nanpy. There are two parts. Putting the slave firmware on the Arduino and installing the library that is needed for the Pi. The following is taken from the How to build and install section of the nanpy page:
First of all, you need to build the firmware and upload it on your Arduino, to do that clone the nanpy-firmware repository on Github or download it from PyPi.
git clone https://github.com/nanpy/nanpy-firmware.git cd nanpy-firmware ./configure.sh
You can now edit Nanpy/cfg.h generated file to configure your Nanpy firmware, selecting the features you want to include and the baud rate.
To build and install Nanpy firmware, copy Nanpy directory under your “sketchbook” directory, start your Arduino IDE, open Sketchbook -> Nanpy and click on “Upload”.
To install Nanpy Python library on your master device just type:
pip install nanpy
Step 2: Hardware
Now we have all of the software set up, its time to wire up the hardware. There are two steps. Wire the Arduino to the relay and wire the relay to the coffee machine start button. The only relay that I had was an 8 channel relay so that's what I used. This could be done with one relay. Also the only relay image I could find was that of a 4 channel relay in the picture below. I hope it doesnt cause too much confusion.
Arduino to Relay
Arduino pin | Wire color | Relay connection
------------- | ------------- | -------------
5V | Red | VCC
GND | Black | GND
Digital pin 8 | Orange | IN1
Relay to Coffee Maker button
Because I didnt know what I was doing at first, with the coffee maker, I actually took apart the whole thing. It was a bit of a pain because the screws on the bottom had a wierd three prong star head. I ended up using a small flat head screw driver that I forced to fit. After I opened it up, I saw that all I needed to do was to pop the button out, which can easily be done with a flat head screw driver.
Relay connection | Wire color | Coffee Maker button connection
------------- | ------------- | -------------
Common | Green | Key
NC | Yellow | GND
You can solder the wires to the back of the button. I dont trust my soldering skills so I simply used a hot glue gun to keep the points connected. This probably isnt the best thing to do but with zero soldering experience, its good enough for now.
Make some coffee
Thats it! Now all we have to do is start the python server on our Raspberry Pi and we can test with Alexa. Start the server by running the command.