does the KY040 work as in spec?

As far as I understood the spec of the KY040, I had assumed for a "constant turning speed" the output of the encoder looks like the following:

Clockwise

Code:


DT  ___/"""\___/"""\___/"""\___/"""\
CLK _/"""\___/"""\___/"""\___/"""\___


Counter clockwise

Code:


DT  _/"""\___/"""\___/"""\___/"""\___
CLK ___/"""\___/"""\___/"""\___/"""\


As far as I get it in the simulation, it looks more like:

Clockwise

Code:


DT  """""\__/"""""""""""""""""""""\__/"""
CLK  """\__/"""""""""""""""""""""\__/"""""


Counter clockwise

Code:


DT  __/""\____________________/""\____
CLK ___/""\____________________/""\__


The high parts (for CW) / low parts (for CCW) are much longer.

So, the edge sequence is right but the "duty cycle" seems strange to me. I thought this should be 50:50 be the sensor wheel.

Background:
I have seen some debouncing algorithm which do fail see any rotation. These were not based on edge or pin change detection, but used periodical probing of DT and CLK.

I didn't write this component, so I don't know much about how it works.
I will have a look.

I refactored this component to match the "real" timing and it's now a generic relative rotary encoder.
By now only the number of step per rotation is configurable.

Done at Rev 961.