Close
0%
0%

Encoder project

hi res absolute hollow shaft optical encoder

Similar projects worth following
Optical sensors are cheap. DSPs are cheap. Surely we can put them together and make an affordable grey code reader.

Previous work

http://tim.cexx.org/?p=613
https://conorpeterson.wordpress.com/2010/06/04/optical-mouse-hacking-part-1/
etc.

The optical mouse sensor above is ~$1.50 USD and has a resolution of 19x19 or 18x18.  Extracting an image takes 0.5s during which the camera cannot move relative to the source.  Surely there is a faster sensor out there?

The goal

20 bits of grey code on either the outside of a cylinder or on a flat disc.  I can make either option work in my robots.

Why

Because right now the best option I have is a magnetic encoder for ~$200 USD: https://www.rls.si/en/aksim-2-off-axis-rotary-absolute-encoder

Your help

What I need from you, hive mind, is a sensor that's <$5 and reads 40xA where A>2 and a very fast SPI read time (several thousand FPS) OR a DSP built in to read the grey code.  Nice to have would be a 3.3v or 5v interface.


View all 3 project logs

Enjoy this project?

Share

Discussions

Simon Merrett wrote 06/09/2019 at 07:21 point

@Dan Royer are you prepared to go analogue initially and eg average readings to get a stable value?

https://botscene.net/2012/10/18/make-a-low-cost-absolute-encoder/

This would work for a ring or disc. 

  Are you sure? yes | no

Dan Royer wrote 06/10/2019 at 19:18 point

I'm even prepared to go RGB sensor and put three colors on the paper.  My hope is that it gives greater precision and no wrap-around error where 0 is mistake briefly for 180.

  Are you sure? yes | no

Simon Merrett wrote 06/10/2019 at 19:38 point

I haven't thought hard enough about this but is there really a wrap around error? If the transition from black to white is perceived as constant across the rotation and is darker at 45 degrees from the black than it is from the white, at what point do you see the error? 

  Are you sure? yes | no

Dan Royer wrote 06/10/2019 at 19:57 point

Imagine a rectangle of paper divided diagonally with one half black.  Rolled around a drum, it forms a saw tooth when the paper loops at 0 degrees.  Two sensors would fix this.  one RGB sensor with three offset saw teeth should also work.

  Are you sure? yes | no

Simon Merrett wrote 06/10/2019 at 21:23 point

But that's not what the grayscale technique in the link I posted produces, is it? Two monochrome sensors would give continuous absolute position without wrap around. 

  Are you sure? yes | no

Dan Royer wrote 06/10/2019 at 22:11 point

correct.  I was considering methods that might use less parts and then be cheaper.

  Are you sure? yes | no

Similar Projects

Does this project spark your interest?

Become a member to follow this project and never miss any updates