A few weeks ago, I wrote a first article on a DIY IoT project called ConnectiCup. In this first article, I explain how I came up with the idea to create a connected mug that tells you when it’s at your preferred drinking temperature. I also go into detail in how I created the prototype in true DIY fashion.
In this second article, I will explain scaling and mass production of the ConnectiCup.
And now? That’s the question we ask ourselves at this point: I had a functional prototype of the connected mug which was not at all produceable. We are certainly not talking about large-scale industry, rather simply to imagine the production of several ConnectiCups to be given, for example, to colleagues at Backelite. In this light, it was unthinkable to create many bases with silicone and corn starch, to drill the holes of each LED and to cut the battery compartment and the charger out by hand. That’s too much manual labor! So then, what to do?
The solution arrived on a platter with our 3D printer. With 3D printing, one can print lots of custom-made objects, objects you would never find in commerce, or at least not in this form. The possibilities are limited by two main factors: the capacity of the printer and your imagination. The only prerequeistie is that one needs to model the desired object with specific software, or to find a ready-made model on the Internet.
For the ConnectiCup, we obviously had to model it. Then the only thing we had to do was to print as many as required and assemble the circuit. Done with the homemade recipes and approximate cutouts made with a cutter and hole punch!
The base modeled in 3D can be broken down into two interlocking pieces, printed starting with pieces of PLA plastic wire: the first is the main piece into which the cup, the battery and the LED ribbon slide, the second houses the rest of the circuit components in specific compartments.
On paper this seemed very simple but in fact the modeling had its difficulties. It even required a lot more time than I expected. The main sticking point was the modeling of the holes which would house the LEDs because they had to be correctly spaced– to the exact centimeter.
A laborious assembly of the connected mug
Once I had printed the first version of the connected mug, some adjustments were necessary, especially related to the space reserved for each component and the space between the LEDs. When the right version was able to be printed (this time in blue), we still had to assemble the components in the base to get a functioning ConnectiCup. This step took me long hours and was very laborious. In fact, the limited space and the space dedicated to each component gave this first assembly a very “custom made” feeling.
Once I had finished my long task, I put the freshly installed connected mug to work. Since it was my lucky day, I hadn’t messed anything up and it worked the first time. The illuminated LEDs even give it a bit of a spaceship look that was really fun! The only minor note was the holder which was slightly too short in relation to the cup and which hindered a good temperature measurement: an adjustment would thus be necessary. In the meantime, I was able to take it off thanks to a small paper support.
Nevertheless, envisioning the production of several ConnectiCups, I couldn’t see making a second assembly like this one. It was too long and bothersome. I had to figure out a way to fix that.
How could I make assembly of the connected mug easier?
The main problem was the large number of wires that crisscrossed each other in every direction. I therefore decided to test a special felt tip marker that lets you draw with conductive ink, the Circuit Marker by AgIC. It would thus be sufficient to replace the required wires with lines drawn with the marker directly on the piece of plastic. Then, all I had to do was to connect these lines to the components with a solder joint and it was done!
To be able to test it, a new piece with tracks dedicated to the marker lines was designed and printed, including a second space for the temperature sensor so that it would be at the right height.
Unfortunately, this marker proved not to be effective except on photographic paper. Impossible to make it hold on the plastic! So I glued a ribbon of photo paper on the piece and covered it with the marker, but that melted immediately on contact with the soldering iron.
I thus turned to the Circuit Scribe, another model that works on more surfaces. As expected, this held on the piece, but when I wanted to apply solder to it, the soldering iron immediately melted the plastic and the marker line with it. After these disastrous results, I decided to forget the idea of the marker and to find another alternative…
The conducting leads solution
I finally opted for an intermediate solution by placing two conducting leads, one for the plus and one for the minus, on one side and the other side of the plastic. Thus, all the wires going to the ground go on the minus path and all the wires carrying the current go to the plus. The two battery wires would be soldered respectively onto the minus and the plus leads. That reduced the crisscrossing of wires and simplified assembly.
For the lead material, copper was the best option. But I had to think “mass production” for the connected mug and I didn’t want to amuse myself cutting out the leads by hand and holding them with glue for each new ConnectiCup…. Searching on the internet, I discovered the existence of copper adhesive tape. You just have to cut the length you want, stick it on directly and you’re done! Plus, the cherry on top is that you can solder directly onto it!
The piece was thus designed again to accommodate these leads. At the time I am writing this, it hasn’t been able to be printed yet and I therefore haven’t been able to test the complete assembly of a new circuit. That will happen very soon!
While awaiting the orders…
To keep myself busy during the wait to receive the markers and the copper adhesive tape, I developed some advances that I initially was holding back for the next version of the connected mug.
I thus created a preferences screen: it is thus possible to vary the intensity of the LEDs, to animate them or not at the time of a notification (with an animation by default), to change the name of the ConnectiCup that appeared on the list of Bluetooth peripherals nearby. The colors of the LEDs are also displayed on the primary screen and the user can delete the pairing with his ConnectiCup if he ever wants to use another model. Finally, the quality of the Bluetooth signal is displayed and it is possible to make your ConnectiCup flash to find it if your office is as messy as mine.
A major part of these functionalities involves Bluetooth communication between the smartphone and the connected mug. This therefore involved the establishment of a small protocol. In fact, since each message transmitted via Bluetooth cannot exceed 20 characters, and since one can only transmit one at a time, I had to simplify everything as much as possible while maintaining something readable. For example, for the display of the LED color in the app, the ConnectiCup sends as many messages as LEDs. Each message contains the number of the LED as well as the code for its color. The app handles one after the other to update the display on the main screen.
I also took the opportunity to conduct a group brainstorming session on the name of the project. In fact, I had no idea at the start and I didn’t want to wait to find a name before beginningl It’s thanks to this big moment of reflection that the name “ConnectiCup” was found and adopted!
Reducing the connected mug costs
When I asked myself the question “how much would a ConnectiCup cost?” and once I did the estimate, I ended up at the amount of 50 euros, which was very expensive compared to the functionalities offered.
The main culprits for this high price were the USB charger and the lithium battery. I hadn’t really considered their prices because I was in the prototype stage when I chose them. Happily, I was able to find a low cost charger, the TP4056. It is equipped with a battery protection circuit and allows one to charge while using the ConnectiCup. I also found a cheaper lithium battery with the same capacity.
Thanks to these new pieces and by buying the other components on cheaper sites, the price was able to be lowered. I was able to reach a total cost of $13.20, being a little less than €12. I haven’t mentioned the price of the plastic for the 3D printing as well as that of the wire necessary for the soldering because they were negligible.
The adventure is far from being finished because I have several ideas for improvements in my head, in particular the addition of a weight sensor that would allow me to add intelligence to the connected mug. For example, while there is still liquid in the cup, a notification could be sent to the user regularly to alert him that his drink is getting cold.
Another subject is that of temperature. During my various tests, I noticed that taking the temperature by direct contact between the analog sensor and the bottom of the cup was slow. Moreover, if the sensor didn’t touch the cup well (just one millimeter of space was enough), the measurement was done badly. This is therefore a very sensitive point and I am counting on doing tests with an infrared sensor that does not require direct contact between the analog sensor and the bottom of the cup, and could be more efficient. I have ordered one, the MLX90614, and I’m patiently awaiting its arrival in my mailbox.
Finally, some other ideas for improvement have been imagined in a more or less precise manner: more advanced energy saving, being able to create one’s own LED animations, make the connected mug ring if you don’t remember in which office you left it, make the ConnectiCups interact with each other. There’s a lot do do!
ConnectiCup is my first connected object project and, despite the difficulties I encountered, I really appreciated it! In the beginning, I didn’t really know where I was going and even if I’d succeed. Today, I am very satisfied with the path I traveled and I’d like to thank all the people who, near or far, have helped me move forward with this project.