The group work
Our group work progress has generally been working really well. At the early stage we experimented a lot, and tried connecting things outside of our exercises. However, one downfall we have had as a group has been, that we have not each individually changed between the tasks. This means that not everyone has tried to make a component inside nodeRED apart from the exercises. We had one time when the work practice was really good: Making the flow diagram, and deciding what the locker should do in different scenarios, made it very clear what tasks should be done. For the next project we will do the flow diagram earlier, define what tasks should be done, and make sure that everyone get to work on all parts of the project.
A new world of things connected to the internet
This project has opened up the world of IoT devices for us. Most interesting is how we could save memory on the nodeMCU, by making cloud computing do most of the tasks for us, by simply communicating with our NodeMCU via MQTT. The nodeRED layout was a very simple way to “program” specific tasks for us. Especially the dashboard functions was amazing to quickly prototype a functional UI, without much knowledge of html and css. Combining it with an MQTT server, made it super easy to send and receive messages (or commands) to the nodeMCU.
It is nice to experiment with making every-day things work on the internet. However as design engineers we must also remember to think about the Raison d’être of our inventions – otherwise they might be chindogu inventions.
The assembly of mechanics and electronics
The construction of the locker was through lots of trial and error. Corners are extremely fragile during assembly, and has costed the lives of a few boxes. We got a tip, that moisturizing the wood before assembly, would make the bendable corners less prone to break. The tip was given post-assembly of the final box, and is therefore something we can take with us to the next project, if bendable laser cut wood still is the thing we want to work with.
We wanted to lessen the resistance in the breadboard, so we could be sure everything would run smoothly, and so no cable would loosen itself during the setup post-assembly. We thought it would be a great idea to solder everything together on the day before delivery, with very little experience in soldering circuits. We got extra cable from the instrument group in building 414. Next time we will ask for something more specific than “cable”, so the instrument group won’t mock us and get a bad impression of Design and Innovation students. In the students workshop in 413 we got a few tips and tricks for creating powerbridges for parallel connections with jumper cables. We got hold of a perfboard which was very helpful on the soldering process.
We were quite excited when we learned that no components was burned in our soldering adventure. With great momentum, we were ready to assemble it. But before we would hide the electronics inside the safe, we had to make sure that it all worked. We had a strange experience with the nodeMCU, that suddenly did not want to connect to our wifi. We re-uploaded the code, and tried to reset it, but nothing worked. We then tried to upload the code on another NodeMCU, and it worked flawlessly. We desoldered the NodeMCU and replaced it with the one that worked. The new NodeMCU now did not want to connect to the wifi. So we tried changing the code, but nothing worked.
The point of no return
At one point we wanted to see if it worked, and plugged in the 12 volt adapter to the nodeMCU. A loud pop went into the room, and we sat back in silence looking at each other. The LEDs had stopped emitting light. It was dead. So we found a third NodeMCU and soldered it onto the board, after cutting loose the old one. And suddenly the wifi was working. So the moral of the story is, sometimes things just don’t work, and if we get stuck, we should try to disassemble everything and try to replace the cables and components.
Chindogu 