With so many cell phones, PDAs, and MP3 players using touch-sensitive keypads, wheels, and screens for user input, regular mechanical pushbuttons are beginning to seem outdated. Even so, mechanical pushbuttons should stay around for a long time, because they cost much less than the controller IC required for a touch sensor.

Suppose you want to improve the look and feel of a product by using a touch sensor, but can’t afford the fancy controller IC. You could then consider a simple four-component circuit which allows a single I/O line from any microcontroller to sense the states of two capacitive touch sensors.

In its standard application, the watchdog timer (U1) uses two capacitors to set the watchdog and reset timeout periods. In this circuit the capacitors are replaced by touch sensors A and B, and the two 4.7-k resistors limit current in the IC’s protection diodes during an ESD event.

WDI is tied low to ensure the watchdog timer always times out, and the Reset output connects to the microcontroller I/O pin. Touch sensor “A” determines the watchdog timer timeout period (tA in the diagram), and touch sensor “B” determines the reset timeout period (tB in the diagram).

Thus, by repeatedly measuring the reset signal’s high and low periods, the microcontroller can determine when either sensor is touched. If a sensor is untouched, the associated period is about 40 μsec. When touched, the sensor signal period rises to 400 μsec or more. Empirical testing can help you choose the timing thresholds that enable firmware code to decide whether a sensor is being touched.

— Eric Schlaepfer

Eric Schlaepfer, Maxim Integrated Products Inc., Sunnyvale, Calif., www.maxim-ic.com