Quadrature Encoders.

Thread Starter

R!f@@

Joined Apr 2, 2009
9,918
I am planing to make a jig with a motor and a QE to enhance my understanding on PIC with QE encoders.

Since I cannot get my hands on a ready made Encoder easily I am planning to make one with the laser cutter and using 3mm plastic.

Need some basic info on it.
I understand the QE outputs a Grey code on phase A and B. And an additional one "Index" can be used to check the Revolution. I know the grey code can also be used to check for RPM.
If I am making why not make one with all so it can be used for all kinda testing, Right.

So here are the Q?
1. Does the phase A and B has to be EXACTLY 90° out of phase and why ?
2. Can I cut the slots at any width ? That is slot that light passes is wider than the one blocking ?
3. Does the slots need to be right angled (0°) to the sensor?
4. Does the sensor A & B need to be close or can I place them slots apart ? Even at 180° apart ?
5. Is there any exact position the Index sensor should be placed ?
6. Does angular position be studied with above setup ?

The lessons will include (but not all in one and sometimes all in one), Revolution counting, Motor speed controlling, Motor direction identification and RPM counting.
Angular position is out for now since it might be complicated.
 

AlbertHall

Joined Jun 4, 2014
12,347
1. No, but the two signals must overlap by a sufficient amount so that the minimum overlap is big enough to allow for the rise/fall times of the sensors; mechanical irregularities (slots not being perfectly cut; eccentricity of the mounting...); the minimum pulse width the PIC can reliably detect; and anything else which might affect the actual widths.
2, 3, 4. Yes with the same restrictions as above.
5. No. It does not have to have any particular relationship to the QE signals. You might want to place it so it has some meaning relative to whatever you are driving - e.g. 'top dead centre'.
6. You can use the QE signals to get angular position at a resolution of the number of slots or four times the number of slots, though the latter won't be quite right if the signals are not 50:50 and in exact quadrature.
 

MaxHeadRoom

Joined Jul 18, 2013
28,702
The most pulses you can get in practice using a simple optical/slot system is around 100 pulses/rev. dependant on disc dia, After this the Moiré effect/system is generally used to increase the resolution to many 1000's of pulses/rev.
The option is there to increase a relatively coarse resolution by using either two pulse edges or all four to increase X2 or X4.

In a CNC application the 1/rev marker is used to set the zero point of the machine, a ordinary L.S. is used to take the axis to the limit where then the axis moves in the opposite direction in slow speed until the marker is seen and the 'Home' registered.
Max.
 
Top