In Hackers, we started work on a short project for a temperature-controlled soldering station. As shown on the whiteboard above, the basic idea is:

• A temperature sensor is connected to an Arduino (analog input)
• A soldering iron, with its tip located near the sensor, is wired to the mains via a relay switch
• The Arduino can control the relay switch to turn the soldering iron on/off
• This is the basis for a simple bang-bang controller: we have a target temperature (e.g. 200 degrees) and a tolerance (e.g. ±10 degrees), then you turn it on if the temperature is below 190 degrees and turn it off if above 210 degrees.

We noted that since there will be space between the soldering iron tip and the temperature sensor, the temperature it will read will be lower than the tip temperature.

As the whiteboard shows, this project also involved discussion of: wiring for the thermocouple; how breadboards are wired; normally-open vs normally-closed relay switches.

One group used a LM35 temperature sensor – here are its specs, and the group found it interesting to see how detailed and useful these data sheets are: http://www.ti.com/product/LM35

The other group used a potentiometer to simulate temperature, so they could test it working over its full range, as shown here:

Below is the code for reading and displaying the temperature. Next steps will be to integrate it with the relay.

// Michael Madden, CoderDojo Athenry
// Reading a temperature sensor.

// The temperature sensor is an LM35 - here are its specs: http://www.ti.com/product/LM35 
// THe temperature range is -55 to 150 degrees celcius.
// Its output voltage pin is connected to an analog input on the Arduino (A1),
// from which we read a value in the range 0-1023, so we convert them.

void setup() {
  // Just needed for print statements to work
  Serial.begin(9600);
}

void loop() {
  // Read the temperature sensor: get a value in range 0-1023
  int val = analogRead(A1);

  // From the analog input, we get a value in the range 0-1023
  // and we need to convert it to a temperature in the range -55 to 150 degrees.
  // The two lines below achieve this in differnet ways and give the same result.
  int temp1 = (val/1023.0)*205 - 55;
  int temp2 = map(val, 0, 1023, -55, 150);

  // Print out the raw value and converted temperature
  Serial.print("Sensor value = ");
  Serial.print(val);
  Serial.print("\tTemperature 1 = "); 
  Serial.print(temp1);
  Serial.print("\tTemperature 2 = ");
  Serial.println(temp2);

  // Wait 100 milliseconds before reading the sensor again
  delay(1000);
}

We had a small but dedicated team of Hackers for our first week back at CoderDojo Athenry last Saturday.

To begin, we started working on 3D printers. 3D printers are a fantastic technology for turning 3D computer models into physical objects. Here are Kevin’s notes on how to set up a 3D printer: 3d-printer-setup (PDF)

Here are the configuration files needed for the Materia 101: https://www.dropbox.com/s/6otj5ok7i00ikds/Slic3r-Materia101-Settings.zip?dl=0

And here also is a diagram Kevin prepared, showing the 3D printing workflow:

In addition to setting up software for the 3D printers on group members’ Windows and Linux machines, we started planning potential projects for this year.

One possible project is to build an 8-bit PC from individual components. It was mentioned that it’s Turing complete, which led to a discussion of some concepts that are named after Alan Turing:

Turing Complete: a computer system is Turing Complete if it has the core features that mean it can run any algorithm. For a modern programming language, this means in practice: memory (variables); decisions (if statements); repetition (loops). However, this does not consider things like how data is input and output (file handling, displays, networking, etc), those secondary capabilities are a lot of what make computers useful.

A nice flip side of Turing Completeness is that if you are learning a new programming language, and you can figure out how to handle variables, decisions and loops in that language, you have mastered all the basics!

The Turing Test is a different concept that Alan Turing came up with, when he was thinking about early concepts of Artificial Intelligence (which was impressive considering how computers barely existed!) Rather than thinking about creating computer intelligence by replicating the functions of the human brain, he imagined an experiment where a human would communicate with either another human or a computer (selected at random); in the conversation, the asker could ask whatever they liked, and the human or computer answerer could try to deceive if they wished. If the experiment is repeated and askers cannot determine accurately whether the answerer is a human or computer, we should conclude that the computer is behaving intelligently.

Incidentally, Google’s recent Duplex demo, where an AI system can phone businesses to make appointments, is arguably an example of an AI system passing the Turing test, asn the people receiving the phone calls thought they were talking to a human: