Nuts and Volts - February 2011

■ FIGURE 4. Breadboard setup before connecting the touchpad.

pin B. 3 which gave me four touch inputs in a row on that side also.

Figure 4 is a photo of my final breadboard setup for the stripboard touchpad, before I connected the two ribbon cable assemblies. In it, you can see the two four-pin, double-ended male headers into which the ribbon cables will be inserted. If you use this approach, don’t forget to make sure that you are using the same four out of five positions on both ends of each ribbon cable. The simplest way to do this is to align the first pin of each male header with the first position on each end of the cable (marked with the red stripe), as shown in the photo of the complete breadboard setup in Figure 5.

TESTING THE COMPLETED MATRIX TOUCHPAD

Hoping to improve the situation, I switched to the touch16 command which was, in fact, somewhat better. Using it, I could reliably detect which column was being pressed, but the row measurements were still unreliable. In addition, my software frequently failed to detect a touch input at all if I pressed the key a little too quickly. I spent several frustrating hours trying various adjustments in the touch16 command’s configuration byte, including several attempts that defined two different configuration bytes for the row and column inputs to see if I could improve the reliability of the row measurements using that approach.

However, I was never able to reliably detect which row was being pressed. I did learn a fair amount about the configuration byte, though. Most importantly, I was able to speed up the touch measurements sufficiently enough to eliminate the problem of quick key-presses being undetected by the software. Changing bits 7-5 of the configuration byte from the default of “000” to “111” easily corrected the problem. If you want to explore these issues on your own, the last test program that I worked with (TouchMatrix Test.bas) is available in the downloads.

Last time, we worked with the 18M2’s touch command. Similarly to the PICAXE ADC commands, there are actually two different touch commands: touch and touch16. As you might suspect, For my first test of the stripboard keypad, I used the touch command to simply measure the values for the eight touch inputs that we are using, and sent them to the terminal window for viewing. I was disappointed when I couldn’t find any reliable differences in the touch measurements, no matter which key I pressed. This was especially true for the row inputs. Of course, that isn’t surprising, since those inputs are connected to the bottom stripboard which means that the row traces (as compared to the column traces) are further away from my finger whenever I touch a key.

touch uses an eight-bit variable, while touch16 uses a 16-bit variable which gives it more precision of measurement. In addition, touch16 also includes an optional configuration byte that can be used to “fine tune” the operation of the touch16 command when necessary. (For details, see the documentation for the touch16 command in Section 2 of the PICAXE manual).

A LITTLE THEORY

■ FIGURE 5. Breadboard setup after connecting the touchpad.

In addition to experimenting with stripboard-based touch inputs, I have also done some web-based research on the topic of designing a PCB version of a touch-sensitive keypad. I didn’t find much useful information until I read a message on the PICAXE forum that mentioned an Application Note (AN1104) published by Microchip Technology, and titled Capacitive Multibutton Configurations. (This is just one more example of how valuable the PICAXE forum can be. If you aren’t already a member, it’s well worth joining!) If you’re interested in reading this application note, go to the Microchip website ( www.microchip.com) and search for “AN1104.”