I’m working on a new exhibit that will be using an Arduino (actually, a Teensy++ 2.0) to talk to an application running on a PC via serial data. The Teensy will be sending one byte to control the application’s behavior. This is an upgrade from an older version where the Teensy just sent keystrokes to the application. The nice thing about sending keystrokes is that it was very easy for anyone to troubleshoot because they could just open Notepad and press some buttons to see if they were sending any output. The bad part was that if a normally closed switch was open, it would just stream characters to the computer, which could make things hard to troubleshoot for some people.
To deal with the troubleshooting issue (which will eventually come up, as it always does) and make it easy for non-technical people to view a serial data stream, I wrote a simple application in Processing that reads the byte and displays the value, along with the status of each physical control of the exhibit.
The exhibit should always have the Teensy plugged into COM3 on the PC, but again, once something leaves the shop we never know what strange things might happen. When the application starts up it will present a dialog showing the COM ports, and asking you to select the correct one. If you select the wrong one it will just display nothing. This should be enough to help troubleshoot things via phone or email.
The trickiest part was the code to choose the COM port. (I know, we don’t call them “COM ports” on Mac OS X, and yes, the application works fine on Mac OS X, that’s another thing I love about Processing.) The code for choosing the COM port came from this forum thread How to let the user select COM (serial) port within a sketch?.
I did have to install Java to get the application to run, but it looks and functions like any other Windows application. Here’s hoping this all works and never has to be used, but is there just in case…
