Let's Make Robots!

Dagu Rover 5 Asymmetrical Encoder Output Problem

I'm having problems with Dagu Rover 5 with asymmetrical encoder outputs.  They do not have a 50% duty cycle and when the two outputs are decoded they don't give the same count going forwards as backwards.  I'm using 74HC14 Schmidt Triggers with a Teensy++ 2.0 and 4x encoder library.  My fix is to count by 1's going forwards and 1.75 backwards.  Is there a way to fix these encoders to have a 50% duty cycle?Dagu Rover 5 Encoder Output

AttachmentSize
IMG_20121021_153839.jpg1.09 MB

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.

This chassis passed QC in 04/12 and has optical encoders.  The circuit board is double sided and looks to have the same circuit on both sides.  The pots are mounted on the bottom side such that you have to remove the board, make an adjustment, screw it back down, close up the case, and run the motors.  Doing so allowed me to get a near perfect 50% duty cycle even though one pot was bottomed out.  The encoders now decode properly and give the same count in both directions. That's probably why they went to Hall Effect though.  I'm going to take it up with Pololu and try to get the chassis swapped out for a new one.

As I said, having a perfect 50% duty cycle is not important. The photo you posted indicated there was nothing wrong with the encoder.

The square waves do not need to be exactly 50% duty cycle to work. The new encoders use an 8-pole magnet and hall-effect sensors. There are no adjustment pots.

Any variance in the 50% duty cycle is due the poles not being 100% identical which is impossible to acheive.All that matters is that one wave form is stable (high or low) while the other is changing state.

The problem will most likely be your code however you do mention using a 74HC14. Why? Please post your wiring diagram and some photos.

I believe that you can adjust the encoders by opening the "legs" of the Rover. I think there is a small pot on the PCB that you can adjust.

Thanks,

The VR1 and VR1 solder pads are there but no potentiometers.  Does anyone have the circuit?  I guess I'll complain to Pololu about it then.

The original encoders were optical and had pots. The new encoders use hall effect sensors. No pots are needed.