Making mead, cider and beer with Scrum4hw

Marcello Semboli
Mar 20, 2018 · 4 min read

I wanted to make mead.
I could find a lot of different receips in Internet to make mead, so, after a lot of reading I started with fixing some basics:
1. I would have to decide the water to honey ratio;
1. I would have to control the temperature;
2. I would have to stir the must several times per day;
3. The product should get more and more transparent along the time.
I’m a mathematician and a computer programmer, my mindset follow me even if I go out of the field of megabytes and enter in the real world. So I wanted to do all operations in a controlled environment.
My first concerns was about the stirring. How could I stir several times per day the must? I am at my workplace for the most part of the day. And how could I control the temperature?

I live in Tuscany, the temperature are between 0 and 12 C in winter and 20–30 C in summer. So all I needed was a simple heater system.
I thought I should build an insulated box to keep the temperature stable. Also I should put a magnetic stirrer inside the box for stirring. Finally I should set some sensor to check the status of the system and put the history in a database.
That was far away my competences and capabilities, I know nothing about electronics but that 30 years old theoretical knowledge I got at the university.
Paolo Sammicheli, a friend of mine told me about Scrum For Hardware and Extreme Manufacturing (http://scrum-hardware.com/), a technique for making things derived from the software development. He also wrote a good book about it, you really should buy it (https://leanpub.com/Scrum-for-Hardware/)
I knew Scrum so I decided to give it a try.
As you can already know, there are three artifacts in Scrum:
- Product Backlog
- Sprint Backlog
- Potentially Shippable Product Increment (PSPI)
I started to writing my backlog as a list features I wanted:
- an insulated box
- stir the liquid
- heat the inside of the box
- write data in a database
- show data in some organized way
I decided that my first PSPI would be an insulated box that could stir a liquid.
The principles of Extreme Manufacturing are the following:
- Optimize for change
- Object-Oriented, Modular Architecture
- Test Driven Development (Red, Green, Refactor)
- Contract-First Design
- Iterate the Design
- Agile Hardware Design Patterns
- Continuous Integration Development
- Continuously Deployed Development
- Scaling Patterns
- Partner Patterns
So I divided my system in three modules: the box, the wiring, the computer. Every modules would have his own interface with the rest of the world. For example: the box was going to have two holes for wires and four little hollows for the sensors. The wiring would interact with the box with sensors and produce output through a serial interface and some messages in ASCII. The computer would have a serial input and a web interface for human interaction.

The advantage of modular architecture is easy to understand: you can change a module in any of his parts, if you respect the interface. You can also trown away a module and rebuild it entirely different, and, if you respect the interface, others module wouldn’t be affected in any way.
Building the box was easy, I’m pretty good with small wooden jobs. I used wood and 3 cm thick polystyrene. I put a PC fan with two magnets on the bottom for magnetic stirring.
I started building a test I could repeat every time I changed something. I used an old charger for smartphone (5V), a jar full of water and a magnetic stir bar. At every change in the box I ran the test to check the stirring was Ok. As an example, I tell you that this technique saved me from a serious problem with woody debris or polystyrene dust that sometime stuck on the PC fan blocking it. I modified the mounting of the fan to avoid the problem.

When the setup was satisfying I declared the first PSPI complete, working and done; so I started with the second iteration.
Skipping boring details, I refined the backlog and I picked up:
- send start and stop commands to the stirring device
- read the temperature
So I added to the wiring module an Arduino and two DS18B20 temperature sensors.

I wrote other software that could test the hardware, over and over, using a Raspberry PI with serial USB connection with the Arduino. Every change I was making in the hardware was tested by the software.
At the end of development, my box was a robust system, that I could turn on pressing a single button and I could configure with my smartphone.
With the first try I made 7 bottles of mead, I don’t know if it is good, because they are still aging.
With this box I can make mead, cider, and beer; it is very versatile and a fun project.

Welcome to a place where words matter. On Medium, smart voices and original ideas take center stage - with no ads in sight. Watch
Follow all the topics you care about, and we’ll deliver the best stories for you to your homepage and inbox. Explore
Get unlimited access to the best stories on Medium — and support writers while you’re at it. Just $5/month. Upgrade