Many motioncontrol engineers prefer optical encoders over other types of position feedback sensors because they are small, relatively rugged, and generate digital output signals that need little conditioning
Most contemporary incremental and absolute optical encoders employ the same operating principles for lightbeam scanning and processing. Light emitted from an LED passes through a rotating disc and a stationary mask to excite a photodiode array. The diode output signal mirrors the pattern on the disc. Incremental encoders have two sets of tracks, one for the count and the other for optional motor commutation. The number of pairs of transparent and opaque areas equally spaced around the rim of the disc relate to the encoder’s resolution.
In comparison, absolute encoder discs consist of several concentric tracks with precisely placed patterns of transparent and opaque segments. Each track generates one bit of a digital word which represents an encoder position. For example, a 12-bit (4,096 parts/rev) encoder has 12 concentric tracks.
Incremental encoders measure position with reference to a start or home position. The most common types provide a digital pulse for each succeeding resolvable position. The pulses are then fed to a high-speed counter in a drive or controller. However, incremental encoders are particularly susceptible to electrical noise. Often, the counter includes the noise pulses along with true position signal pulses. The result is an incorrect position referenced to the start pulse. The stray counts accumulate and can be eliminated only by a homing routine.
Incremental encoders come in two types, single output and quadrature. A single output encoder, often called a tachometer, is normally used in systems that rotate in one direction only and require simple position or velocity information. Velocity data are generated by measuring the time interval between pulses or the number of pulses within a given time period.
Quadrature encoders have dual channels, called A and B, phased 90 electrical degrees apart. The two output signals determine the direction of rotation by detecting the leading or lagging signal in their phase relationship. Also, resolution can be boosted by multiplying the output pulses by some factor. In a dual-channel encoder, a four times increase in resolution is possible by externally counting the rising and falling edges of each channel. For example, a 5,000 pulse/rev quadrature encoder can generate 20,000 pulses/rev with this technique.
In addition to providing accurate pulses, precise position feedback depends on high immunity to the noise-producing false signals. Incremental encoders are especially susceptible to noise when their cable is near large motors or switching gear. One solution for eliminating or reducing noise is providing encoders with complementary outputs. As shown, complimentary signals generate two simultaneous outputs. As channel A goes high, its complement channel B goes low. If this doesn’t happen, the signal is assumed to be influenced by electrical noise and is ignored.
Absolute encoders, on the other hand, generate a unique code word for every resolvable shaft angle without regard to home position. Positioning is precisely determined after the motion system is powered up or after the system is moved when powered down. Absolute encoders don’t require a homing sequence on startup, at routine intervals during operation, or during power and equipment failures. This is critical in safety related applications or when processing expensive raw material that could be damaged by an unplanned homing routine. At a minimum, all axes can reduce cycling time by eliminating this housekeeping operation.
Moreover, absolute encoders don’t accumulate errors. When noise corrupts a position value, only that particular transmission is affected. The next position value cannot be affected by the same noise signal, and previous errors are corrected in the subsequent reading. Also, drive systems can use the absolute value from the feedback device to generate motor commutation signals. This eliminates the need for a separate generator and simplifies wiring.
Programmable absolute encoders have a microprocessor interface module that lets system designers enter several operating parameters including resolution, offset values, output codes, and direction of rotation through a serial interface. This versatility allows a single style of encoder to be stocked and programmed to meet the unique requirements of a job. The encoder can also be easily reconfigured when the application changes.
Single-turn encoders are best suited for short travel motion control applications where position verification is required within a single turn of the encoder shaft. A multiturn encoder is recommended for applications involving multiple revolutions. A multiturn encoder consists of a series of discs connected to the basic high-resolution disc through a system of gears. The additional discs count turns so that position data are available over multiple revolutions.
Encoders may be mounted on motors to close internal feedback loops. Or they can be mounted on loads to provide external feedback directly to a motion controller or serve as an outer feedback loop in a dual-loop drive system. Feedback system performance and functions are dictated by the application. In the main feedback loop that runs the motor, for example, the drive requires dynamic velocity and position feedback as well as commutation information. The outer feedback loop, which in most cases corrects for mechanical backlash in the drive system, often only provides position feedback that can be compared with the inner loop. Encoders mounted directly on the load may provide both speed and position feedback, or in many cases, position feedback only.