Methods of proximity sensing
A number of technologies exist to carry out proximity sensing including, resistive, inductive, capacitive, optical, acoustic and visual methods. Each of these technologies has its advantages and disadvantages. Picking one against the other depends on the specific application, cost and usability. This article will discuss how to implement a capacitive proximity sensor.

Figure 1: Capacitive Sensing Element

To understand how capacitive proximity sensing works, consider the cross section of a capacitive sensing element as shown (Figure 1 above). The conductive copper areas and conductive sensors are below an overlay material. Two conductive elements in close proximity of each other create a capacitance called parasitic capacitance.

The parasitic capacitance is created by the coupling of the sensor pad and the ground plane. The parasitic capacitance is usually in the order of 10pF to 300 pF. The proximity of the sensor and the ground planes also creates a fringe electric field that passes through the overlay.

When a conductive object like the human finger is brought near the fringing electric fields (Figure 2, below), it adds conductive surface area to the capacitive system. This change in the overall capacitance of the system is used to detect the presence of the finger near the capacitive sensor element.

Figure 2: When finger is brought into proximity, there is a change in overall capacitance of sensor

Detecting and measuring the capacitance change
The accuracy and reliability of the proximity sensor depends on accurate measurement of the changes in the capacitance of the system. A number of methods exist for the same. The commonly used ones are Charge Transfer, Successive Approximation, Sigma-Delta and Mutual Capacitance measurements, each having its advantages and disadvantages.

The more commonly adopted techniques on present capacitive sensing ICs are Successive Approximation (CSA) and Sigma-Delta (CSD). Both methods use switched capacitor circuitry and use an external modification capacitor (CMod).

In the CSA method (Figure 3, below), the switched capacitor network charges CMod. The voltage on CMod is then routed through a low-pass filter into a comparator, where it is compared to a reference voltage. A counter clocked by an oscillator is gated by the output of the comparator, the output of which is processed to determine the status of the sensor. CSA requires very few external components. It is also not affected by power-supply transients.

Figure 3: CSA block diagram

In the CSD method (Figure 4, below), the switched capacitor input stage contains the sensor capacitor CX. The switched capacitor network is between VDD and the voltage at CMod. A Pseudo-Random Generator controls the switching frequency of the switched capacitor network. CMod continuously charges and discharges.

When the comparator trips, the bleed resistor switch closes, discharging CMod until a new value is stored in the synchronization latch. The bit stream output of the latch is then 'ANDed' with the PWM and enables a counter.

Figure 4: CSD block diagram

The output of the counter is processed to determine the status of the sensor. CSD is ideally suited for white goods, industrial and automotive applications since it is least susceptible to electromagnetic interference and radiated emissions.