Arduino Teaches Old Coder New Tricks
Because some of the components I was using do not exist in the built-in library, I scoured the Internet for contributed symbols, and in a few cases, I had to design my own symbols. A good source for contributed symbols and footprints is http://gedasymbols.org. For creating your own symbols, see David Weber's Online Symbol Creation Tool at http://EmbeddedToolBox.com. Symbols actually are text files. Figure 3 illustrates a symbol along with a portion of the text file used to draw it. Symbol files are not just an image. They also hold important pin definitions and the name of the footprint file that the gEDA PCB program ultimately will use to represent the component on the circuit board.
Figure 3. Symbol Example 1
A gEDA schematic is a text file interpreted for GUI presentation by gschem, but it also serves as the source for gEDA's PCB program. An intermediary helper program named gsch2pcb is used to prepare the schematic file for use as input to the PCB program. While xgsch2pcb is a GUI version of gsch2pcb for gsch2pcb, I use the gsch2pcb command-line version. For example, given the schematic file vt100lcd84.sch as an input, gsch2pcb creates vt100lcd84.pcb, vt100lcd84.net and vt100lcd84.cmd, all necessary files for PCB creation. gsch2pcb also displays important instructions as part of its command-line text output. To make the process a little easier, I use a file named "project" in the project folder for the current design. Figure 4 shows my project folder for the vt100lcd84 project, the "project" file and the command line with the gsch2pcb command just before execution.
Figure 4. Example of gsch2pcb Project File
It is worth noting that the gEDA suite includes circuit simulation capability (SPICE), enabling virtual design testing. I did not use SPICE with my VT100-LCD project, but see the Resources for this article if you are interested.
Now that I had the circuitry designed for the project, it was time for the software. I wrote the software as a simple state machine that parses each character received on a character-by-character basis, meaning that there is no buffer. Characters are handled differently based upon the current state of the machine. If the state is NOTSPECIAL, the character simply is passed to the LCD screen for display. However, if the state is GOTESCAPE, GOTBRACKET or INNUM, the character is processed further. For example, if the state is GOTBRACKET, both an escape and left-bracket character have been received previously, and the current character must be parsed in that context. For illustration, the VT100 sequence for Screen-Clear is \033[2J, and if the current character being parsed was the 2, the state would be GOTBRACKET, and the next state would be INNUM (number collection).
This method of parsing has the advantage of simplicity, which is suitable for the limited-capacity microcontrollers but with the disadvantage of not being able to scroll the screen due to the absence of a buffer holding a copy what is on the screen. See Resources for a copy of the software source. I used Arduino libraries to build the code. Although the source can be compiled using the Arduino IDE, I used Linux make. Using the Arduino libraries makes the project extremely easy to build. Most of the drudgery of low-level code and the bootloader is hidden away within the Arduino libraries, which freed me to focus solely on my project. Even main() is hidden away such that Arduino code contains two required routines: setup() and loop(). Main actually does exist deep in the Arduino directory structure in ~/arduino/arduino-1.0/hardware/arduino/cores/arduino/main.cpp and is automatically linked in at compile time.
Edward Comer is retired from the telecommunications industry, having worked for the real AT&T, BellSouth and Numerex Corp during a 30-year career.