Week Five

This week's goals are laid out as follows: model capacitive sensing with LEDs, purchase the acrylic board and design as appropriate, draw the all encompassing circuit diagram, and program the Arduino. We divided these goals among ourselves. Gabrielle drew the circuit diagram, Amy investigated capacitive sensing, and Mercedes ordered the acrylic board and created a CAD model for laser cutting.

Three opaque white acrylic boards (5" x 13" x 1/8") were ordered: one to be cut for the top, one for the base, and another to be cut in pieces for the sides of the keyboard. A screenshot of the CAD model is below. Squares, to be cut to insert the keyswitches securely, were drawn on the model. Those squares are the only lines to be actually cut; all other lines are for spacing purposes.

Below are some images of the circuit diagram. This diagram should allow for faster soldering and wiring, as well as for general organization. The diagram does not have the required elements for capacitive sensing yet.

An overall view of the circuit diagram. The arrangement of rows and columns, corresponding with keys, are visible.

A close-up view of one section of the diagram. The many column wires are visible.

A second close-up view of a section near the bottom of the diagram. The arrangement for one of the spacebar's LEDs is visible.

In addition, a preliminary Arduino program was written for keyboard control. A keypad library was downloaded, and Arduino-provided examples were used as models.

Trying to calculate the capacitance of the touch sensors on the home row keys proved hard to do with half the capacitor functionally being a human finger. Research showed conflicting views on the kappa constant of human skin and how deep the functional dielectric properties of human skin go. After some cursory calculations that did not appear to yield accurate results the decision was made to place the plastic key labels over the conductive paint so that the dielectric is plastic, a reliable easily calculated quantity. The greatest possible contact area is 1/8 square inch; an assumption was made that the typist could hover over the home row and still be typing using all ten  fingers giving a separation of 1/32". This assumption was made because the plastic key labels claim to have a width of "80 mkn" a unit that did not appear anywhere in a way that it could be converted to a more familiar unit of length. Lastly the dielectric constant for vinyl is listed as being between 2.8 and 4.5 yielding a range of possible capacitance values.
C=.2249*A*K/d
C=.2249*0.125*2.8/0.03125 = 2.52F
C=.2249*0.125*4.5/0.03125 = 4.05F
These values appear to be much larger than what was expected so when the vinyl labels arrive the capacitance will be measured directly. The importance of knowing the capacitance is to calculate the RC time constant for the purposes of programming with it. Good news update: We don't actually need this information.