Originally, my planning for this week included:
Programming some test applications for the new touch sensors
(if there) Putting multiple FSRs and resistors in the circuit
However, there were several problems that I faced:
- Logistic problems
There were not a lot of shops inside Paris selling cheap small FSRs. The Lextronic shop seemed too far away, and ordering online didn’t work as the sensors could arrive too late.
- Programmatic problems
As mentioned in last week’s blog: so far I had not managed to send the values from NodeJS to a web interface.
This is why I could do neither test applications and putting more FSRs there. So what I decided to do is work on other important stuff:
How to make my circuit smaller so that I can put it in theory on the back of my phone?
Find smaller Arduino
It seemed clear that the Arduino Uno would be too big for the backside for my iPhone 5, and I had to get rid of the breadboard as well.
Luckily, I was offered the Arduino Nano, the size of a USB-stick and much thinner. I quickly read some of the info on Arduino Nano on the Arduino community website. It turned out I could simply recreate the circuit on there, but for that I had to learn how to solder.
Find new resistor
Soldering five times 10.000 Ω resistors would make no sense, so I first had to find a 50.000 Ω resistor to replace it.
In one of the shelves of the cupboard in the Fab lab, there was a box full of maps of resistors. I checked the Arduino book to see how a 50.000 Ω resistor would look like. I figured out the colored stripes combination, but searching for that would take way too much time. Luckily on the folders the exact resistance value was mentioned.
I did not find a 50.000 Ω resistor, but I did find resistors of 56.000 Ω and 47.000 Ω. I decided to put them in the calculation (see last blog) and check what voltage would go out into the analog pin:
Vout = 56.000 / (56.000 + 200) * 5 = 4.98
Vout = 56.000 / (56.000 + 100.000) * 5 = 1.67
Vout = 56.000 / infinity * 5 = 0
Vout = 47.000 / (47.000 + 200) * 5 = 4.98
Vout = 47.000 / (47.000 + 100.000) * 5 = 1.60
Vout = 47.000 / infinity * 5 = 0
Since both Vouts were approaching 5V and 0V really close, it didn’t really matter. In the end I picked the 47.000 Ω resistor because it was closer to the original 50K.
Then, I had to learn how to solder wires to each other and to Arduino. I picked a thin, flexible wire. The colour did not really matter, but we used red and black for power and ground wires like traditional.
First I had to estimate how long each wire had to be. In order to do that, I traced the outline of the back of my phone and sketched the rough layout of the Arduino (lower half) and the FSR (upper half) on it. From there it was easy to determine the approximate length of each of the wires, so I cut the wires with a pair of pincers.
For the soldering itself, I did as follows:
I removed the end of each of the wire covers with the wire stripper. Then I was told it was good to already pre-solder (« tin ») the wire-end alone. This helped putting the small sub-wires together, making the actual soldering much easier. When I did this, I noticed that the material quickly turned a bit shiny and rope-alike.
- Soldering to other wires
I turned on the machine, took the hot part out (right hand), put that firmly on the wires that I wanted to connect and added the soldering wire (left hand) for about 4 seconds. After good soldering it became apparent that they were stuck to each other.
- Soldering to Arduino
The same procedure holds for the Arduino, but now I had to connect the circular part of an Arduino input to the wire-end instead. After good soldering in this case I could see a ball-shape appearing on top of the circle.
In the beginning I found it quite tricky to get used to the whole process, especially the order of the movements, the time to put down each element and also taking security measurements like holding the soldering wire far from your hand and putting the heating element back in its case. Also the sizes of all elements were very small. For once I could imagine why robots were much better circuit creators!
In the end of the project, it turned out that the wires that I had estimated were still a bit too long, folding them in a bad angle. This caused the output wire to break, and I had to solder it better, again.