Easy e-field imaging

Ron DeLong
Senior Applications and Systems Engineer
Analog Products Division, Semiconductor Products Sector
Motorola Inc.
Tempe, Ariz.
www.motorola.com

The Motorola MC33794, together with a microcontroller, simplifies electric-field imaging. The chip supports up to nine electrodes and has built-in watchdog and power-on-reset timers.

 

The MC33794 in a typical application together with a microcontroller.

Electric-field (e-field) imaging is not new, but implementing and using it just got easier. A single chip now replaces nine ICs, seven discrete transistors, and more than 70 passive components.

The chip, the MC33794, generates an analog sine wave producing a low-distortion signal. An external resistor sets the actual frequency, typically around 120 kHz, but can range between 60 and 180 kHz. The low distortion keeps the harmonic content low to avoid interference with AM broadcast receivers.

The IC works with a microcontroller (MCU) that integrates an on-chip analog-to-digital converter. This ac signal has a voltage swing of 5 to 8 V peak-to-peak inside the IC and is routed through an internal resistor to an analog switch. The switch routes the signal to one of eleven pins. Up to nine of the pins connect to electrodes for sensing, with two connected to reference capacitors that allow using measurement-correction firmware in the system MCU.

Electrodes connected to pins create a voltage drop across an internal resistor. The drop is due to the current flow from the electrodes to the circuit ground-return path. Objects in the electrode field affect the magnitude of the current, reducing the amplitude of the ac voltage at the pin. A detector circuit in the IC converts the ac amplitude to a dc level. An external capacitor removes ripple from the detection process and sets the response time of the output, typically from 10 to 250 msec.

Electric-field imaging starts with the electric field generated by a voltage potential between two conductors. The electric field between the conductors exists between them in semicircular paths. Objects in the field change its magnitude which is sensed by the chip. With more than two electrodes to map an object, the chip determines more than just an object's proximity or presence.

Magnetic and inductive sensing are based on the same technology and differ from e-field sensing and imaging in that magnetic and inductive sensing rely on magnetic flux lines created around a conductor. Magnetic and inductive sensors easily detect materials with a high-permeability such as iron but are little affected by low-permeability (nonferrous) materials such as aluminum. They're also not directly affected by material conductivity or dielectric constant. So for instance, the human body is difficult to detect with magnetic or inductive sensing, but easy to detect with e-field imaging.

There are many uses for e-field imaging. Engineers at the MIT Media Labs, for instance, have developed an application that produces 3D images of a human hand from a 2D array of electrodes. With enough data-processing muscle, an image of the hand can be displayed to show orientation and position of the hand and fingers.Another application comes from the auto industry. Until recently, a major problem with side air bags has been that while they are beneficial and safe for an adult sitting upright in the seat, they can injure a child or adult when crouched or sitting at an odd angle. This is especially true when the passenger's head is near the air bag as it fires. The sheer force of a deploying air bag can injure the person's head and neck.

One solution places foam-conductive electrodes under the seat upholstery and lets electric-field imaging determine the passenger's size and position. Electrodes placed at intervals on the seat back can determine the person's height. Another electrode in the air bag can determine the proximity of any part of the individual's body to the air bag. These combined factors would determine whether to allow or suppress the firing of the air bag during an accident. Earlier designs used discrete components, which not only increased size but also cost more than newer semiconductor technology now available.