Making Electronic Sound With Conductive Plaster

Following blorgggg's project on conductive silicone circuit, I decided to venture on my own experiment with carbon fiber. Turns out, a shape cast out of carbon-fiber-infused plaster can also be used as a variable resistor! With a few copper rod and the a few quick programming, you will be able to use your conductive plaster form as a sensor which, in this particular example, will be used to generate sound.

The application of this experimental form goes well beyond making electronic sound itself. I share this project in the hope of expanding the possibility of circuitry. Electronics do not always have to live within a neat and sleek container; they can also be thought of to be within sculptures, materials, forms, and everyday objects--and we will enter this project with the mindset of creating an alternative to knobs, inlets, or buttons. We will be creating a structure for circuitry that is uncertain and full of surprises. And so without further ado, here are some of the stuff you'll need to prepare.

Things you will need for casting:

• Dust mask (very important for the longevity of your lungs!!!)
• Any type of casting mold. I use a mold that I made using Smooth-On silicone, of an enlarged LED shape. If you do not have any, you can get a pre-existing mold (if you're not too concerned about shapes, even a cupcake/ice mold would do) or look through different how-to tutorials.
• Plaster (any kind, but I prefer USG Hydrocal because they are strong and durable)
• 2 measuring cups (1 quart and 8 oz.)
• Mixing sticks
• Mixed chopped carbon fiber (available on eBay)
• Denatured alcohol fuel (you will be able to find it at a supply store)

Things you will need to make the circuitry:

• Arduino Uno/Nano and their corresponding USB cables
• Solderless breadboard
• Multimeter
• Copper rod (1/16" - 1/8") and a drill with a drill bit of the same thickness as the rod
• Multicolored wires (I use 22 gauge Striveday silicone wire because of their elasticity)
• 22k Resistors
• Electric tape

Programs you will need on your computer:

• Arduino IDE
• Pd-Extended (a sound programming language) and the convert.zip folder (to be used later)

Let's begin!

The best way to measure the volume of the cast is by filling the mold up with water, and then pouring that water onto a measuring container. In my case, I found out that my form has a volume of roughly 11 oz. With this number, I will check the data sheet of my plaster and find out how much water and plaster I will need. The ratio is different with each plaster product, so do double check. In the case of using USG Hydrocal to cast my form, I need 8 oz. of water and 11 oz. of plaster.

Fill one quart cup with the amount of water that you need, and another with the corresponding amount of plaster.

The more carbon fiber put into your plaster, the more conductive the plaster will be. At a certain point however, a high concentration of carbon fiber will interfere with the structural integrity of the plaster, and it will cause difficulties in mixing. For 11 oz. of plaster, I figured that infusing 1.5 teaspoon of carbon fiber is enough to make it conductive even after the plaster dries. So I suggest using around 1.5 to 2 teaspoon of carbon fiber / 10 oz. of plaster

Put this amount of carbon fiber in the 8 oz. measuring cup, and submerge it lightly with denatured alcohol. Take a mixing stick and whisk the carbon fiber until there are no visible chops left - it should look pretty close to the image above. Pour out the excess alcohol, and let it sit for a second (but not until the alcohol dries, since the carbon fiber will stick to itself again!)

Dump the carbon fiber into the one quart container with water in it.

Do not forget to wear a dust mask!

Start sprinkling plaster powder into the carbon-fiber filled water, steadily while constantly stirring. This will ensure that the carbon fiber is constantly being dispersed inside the water. Keep a lookout for lumps of plasters and chunks of carbon fibers, and break them apart on the wall of the container with the mixing stick. Continue doing so until you can feel a little bit of a resistance while mixing, and the mixture begins to have a milkshake-like consistency. As this happens, make sure that there are no more clustered carbon fibers.

There are two conditions to look out for:

1. Once the water is saturated with plaster, the additional plaster that that is sprinkled will form craters and islands on the surface. Continue to add plaster until the islands of plaster stop absorbing water / forming craters.
2. As you stir the mixture, the strands of carbon fiber should move in a flow pattern that follows the direction of the stir.

Once these two conditions are fulfilled, pour the plaster into the mold vigorously. This will ensure that the carbon fiber strands end up intersecting each other, therefore forming a connection of conductivity.

While waiting for the plaster to cure, you can start making the copper connector. There are two types of connectors:

1. The one that goes from the breadboard and measure values

Cut a length of cable, around 12"-18". Strip 2" of cable on one end, and about 1/2" on the other. Splay and spread the strands of wire on the 2" end, and twist them around the copper rod, going up to about halfway of its length. Solder on and around the strands of wire, ensuring that the wire is pretty firmly secured on to the rod. After letting it cool for about 2 minutes, wrap the soldered part with electrical tape. Twist the other end firmly such that it can be inserted into the breadboard. (Optional: you can also solder the shorter end to a piece of solid wire / jumper wire, since they are more friendly to solderless-breadboard)

For this tutorial, I recommend making 4 of these connectors, since the code I provided is made for 4 connectors.

2. The one that connects different plaster forms

Basically the same as above, except this time both ends will have copper rod on it. 2 or 3 of these connectors would do.

It is a good idea to have different colored cables, since the tangle of cables might be rather confusing later on.

After about an hour and a half, the plaster form should already cure. If the exposed surface of the cast is warm and solid, the plaster cast is ready to be demolded. If it is still a little soft and damp, give it another 15-30 minutes of waiting.

After that, drill a few holes with the drill bit that is no more than 1 1/2" deep on your forms, spreading them around pretty evenly. If you do not fancy drilling holes into the form, don't worry! The whole surface of the cast is conductive and therefore just by brushing, the copper connectors can still conduct electricity. (You can even use your own body and its resistance to conduct the electricity, and again no worry! We will make sure that the running electricity is within the range of being body safe) However, a hole provides a nifty resting hole for the connectors, and therefore you wouldn't have to worry about having to hold on to many connectors at once.

The way the circuit works is basically the same as about any variable resistors. You will basically need 3 jumper wires, a 22k ohm resistor, and the two copper connectors. You can play around with different resistors later on to change the value that you will obtain. However, I found 22k ohm to produce the most versatile range of values.

The above diagram only shows how to make one connection that reads one value. However, you can add more connectors depending on the number of analog inputs that you have on your board (I like to use the Nano because it is compact and it has 8 analog inputs). You will only need one copper connector going to GND.

WARNING: Only use a regulated 5V power supply for the input! Meddling with a power supply higher than that may cause shock, especially since we are dealing with open circuitry.

After you have your circuit set up, connect your Uno/Nano to your computer via their corresponding USB cables. upload this code to your board.

After uploading, take note of the port number from which you upload your sketch. You can find this out in the Arduino IDE, through Tools -> Port.

<p>float value1, value2, value3, value4; // you can add more of these values depending on how many connectors you have</p><p>void setup() {
  Serial.begin(9600);
}</p><p>void loop() {
  value1 = 1024 - analogRead(A0);
  value2 = 1024 - analogRead(A1);
  value3 = 1024 - analogRead(A2);
  value4 = 1024 - analogRead(A3);</p><p>// add more / delete some depending on the number of connectors
  Serial.print(value1); Serial.print("_");
  Serial.print(value2); Serial.print("_");
  Serial.print(value3); Serial.print("_");
  Serial.println(value4);</p><p>// PureData reads value that is separated by an underscore, so make sure you add a Serial.print("_") after each one, ending the list with a Serial.println(valueX)</p><p>}</p>

Install PureData Extended, and unzip the attached folder. Open the patch named soundtest, and you will see a line of nodes on the PureData IDE. Click Edit, and check Edit Mode.

Click the top message object that says "Open 8" and change the number 8 to the number of your port.

If you have more / less than 4 connectors, add / remove a number of "f" from the box that says unpack. After doing so, you can play around with the algorithmic structure of the sound. I would recommend looking into more tutorials of PureData, which are thorough, informative, and well documented -- and the best part is, it can be easily found in their own IDE, through Help -> Pd Help Browser... .

Uncheck Edit Mode and click on this object. (Note: you will not be able to upload any sketch to your board when the comport serial is open in PureData). A stream of value should be appearing, changing the value on the grey box that used to say 0. Connect / brush your copper connector on one, or even multiple plaster form, and now you are able to generate sound!

The question of what's next is a vast, and open-ended question. My experimentation with conductive plaster is only at its premature stage, but I am certainly hoping that other makers will be engaged in answering this question, not only technically, but also critically. What if and what would happen if our walls are conductive? What if and what would happen if the values obtained from these plasters are used for data visualization instead? What if and what would happen if a plaster object can be a new form of data cryptography? What if technology is not just limited to the purview of giant companies, to the containment of manufactured plastic and CNC milled aluminium container? I'm excited for all these possibilities, and I'm excited to see how others makers will riff off this project, creating something new, unexpected, and beautifully, and necessarily imaginative.