Encoder outputs and determining direction！

All of the encoder designs discussed are incremental types, meaning they indicate relative position of the shaft, but not its absolute position. Thus, at "power-up" the system will not know the angle or position of the shaft. The alternative approach is to use an absolute encoder, a more complex component that reports angle as soon as power is applied. Absolute encoders are more commonly needed on shafts that do not rotate a full circle or more, such as instrument bearings used to determine the angle of tilt in a surveyor transit. However, these are not conventional motor-rotation situations, and even on these instruments, the absolute encoders face competition from low-cost MEMS accelerometers.

Optical encoder vendors have fortunately standardized on encoder output-signal formats, which eases choosing among the different technologies. The "raw" outputs of the encoder are the primary or in-phase signal, often called the A signal, and a quadrature (90⁰) signal offset from the primary track, often called the B signal.

For rotation in one direction, A leads B by 90⁰, for counter-rotation, B leads A by 90⁰. Thus, by looking at the two signals and their relative phase, the direction of rotation is determined. The encoder's associated electronics produce "count-up" pulses or "count-down" pulses that the system controller can then use to determine relative position, speed and even acceleration (the latter requires more real-time processing, and may be sensitive to signal jitter).

Fig. 5: An optical encoder generates two signals in quadrature; by looking at their relative phase difference, the system can determine the direction of rotation. Not shown is the index pulse, if any. (Source: Robotoid, Robot Builder's Bonanza, 4th Edition - Application Notes & Bonus Projects; released under Creative Commons 3.0 SA License)

Increasingly, motor-shaft sensing designs that may have used absolute encoders in the past now use incremental encoders with the third index channel, to determine the shaft position on power-up if needed. This works because knowing the absolute position on start-up is not important in many applications, while the relative position and motion information is critical and the once-per-revolution index pulse is sufficient.

While the encoder's basic output A, B and index signal functions are standardized, these signals are provided at different levels and compatibilities depending on the specific encoder model. Users can choose among TTL, CMOS, single-ended, and differential A, B, and index signals, to match the interface requirements of the electronics that connect to the encoder output. Most vendors offer many of their encoders with a choice of interface options, so the user can first pick a unit with the needed encoding performance and then select the appropriate electrical interface.

Previous：How are Encoders Controlled?
Next： What’s the difference between single and multi-turn rotary encoders?