Rotary encoder: pulse per 360 deg rotation and detent

I've only used rotary encoders that have the same number of detents and pulses per 360 degree of rotation, such as 18 pulses per 360 degree of revolution and 18 detents.

I am used to the fact that CH-A and B are both HIGH for the encoder in a detent, and then CH-A LOW followed by CH-B LOW and then CH-A returns to HIGH and CH-B then returns to HIGH. This means the encoder is turned in a certain direction on detent.

I've recently found some encoders that have twice the detent as they have pulses, such as PEC11R-4230F-S0012 from digikey:

The spec sheet claims this model has 24 detents and 12 pulses. Does this mean that the encoder could sit on a detent location that is one CH-A LOW then CH-B LOW, without either channel returning to HIGH, indefinitely?


Rotary encoders of this type output two voltages or logic signals 'A' and 'B' in the following repeating sequence:

11 01 00 10 for clockwise rotation CW
11 10 00 01 for counter-clockwise rotation CCW

The sequence consists of voltage levels or logic 'states' which accord with the current position of the shaft and are constant while the shaft remains in that position.

In encoder literature a 'pulse' refers to one complete cycle of the sequence, for reasons that aren't immediately obvious to me, which seems rather confusing.

The Bournes PEC11R-4230F-S0012 encoder you refer to produces, as you say, 12 pulses per full rotation and has 24 detents.

The Bournes datasheet and the limited incremental encoder application guidance on the Bournes website nowhere that I can see actually define the meaning of a 'pulse'. The datasheet does include on the output waveform diagram an unexplained 'D' which can reasonably be taken to correspond to the output at a detent position, 11 as you indicate from your past experience. But there is no indication of the output at the detent positions for variants of the encoder where the number of detents is different from the number of pulses.

In the case where the number of detents is twice the number of 'pulse' per evolution, it seems logical to me that the output states at the detents will be alternately 11 and 00 .

Does that seem reasonable?

Thanks. That was what I was thinking about. I've ordered one such encoder. I'll report back in a few days.