Rotary encoders measure motion by reading a pattern on a rotating disk. Contact encoders read the disk with brushes or finger sensors. Noncontact encoders rely on magnetic, optical, capacitive or inductive technology. Of these, the most common are magnetic and optical.

A magnetic encoder has a magnetoresistive sensor that detects changes in magnetic flux. The disk is magnetically coded, and the sensor interprets the code as a series of on and off states. Magnetic encoders offer good resolution. They can operate in harsh conditions and require low power for operation. However, they cannot achieve very high speeds.

In an optical encoder, light is projected through thin slits in the disk. The disk can be glass or plastic with thin lines etched into a coating or plating, or the disk can be metal with etchings through it. The light source is a lamp or LED, but lasers are someArial used in applications requiring very high resolutions. A photo-receptor on the opposite side of the disk detects the light and converts it to electrical signals.

Optical encoders are highly accurate and provide high resolutions. Some offer more than 1 million counts per rotation.

"Magnetic encoders are more forgiving in dirty environments, but you get better resolution and a cleaner signal from optical en-coders," says Tracey Howard, marketing manager for Encoder Products Inc. (Sandpoint, ID).

Whether contact or noncontact, rotary encoders can provide either incremental or absolute positional information.

The codes on an incremental encoder’s disk are uniformly sized and evenly spaced, says Scott Orlosky, manager for marketing and international sales at BEI Technologies Industrial Encoder Div. (Goleta, CA). As the disk turns, the encoder transmits one pulse for each spot on the disk. The controller counts the pulses to determine the shaft’s speed or its position relative to another position. Incremental encoders provide more resolution at a lower cost than absolute encoders. However, they cannot determine their location at start-up. Instead, they must run a homing sequence to find a reference pulse, and begin from there.

Absolute encoders provide a unique code for each angular position within their range of resolution. The advantage of absolute encoders is that they always know their position, even after a power loss. But, they are more complex and expensive than incremental encoders.

"Absolute encoders are good if you need to know exact positions, but they’re not good for measuring speed or acceleration," says Howard. "Incremental encoders are great for measuring speed and acceleration, and they’re OK for positioning, as long as you don’t lose power."

Absolute encoders come in single-turn and multi-turn versions. A single-turn encoder cannot count how many rotations a shaft has completed, explains Jeremy Jones, product manager for encoders and display devices at Baumer Electric (Southington, CT). Because each angular position has its own code, the encoder resets to zero after each full rotation. A multi-turn encoder has a system of gears or magnetic sensors to measure rotations of more than 360 degrees.