Project Logs

Collapse

• The hopper saga, shaking some skittles

  Gustavo Laureano • 08/14/2023 at 17:55 • 0 comments

  As you already know from my previous update my focus is now in developing a Skittles hopper that doesn't jam, after some thinking about how hard would it be to transfer the movement from the disc to inside the hopper (my main design goal is to only use 1 motor, remember?)  I came to the conclusion that it is easier to use a servo with extended angle range (270 degrees of movement), using the 1/4 of turn behind the color sensor as "stirring" zone, so I installed this servo:

  Which goes well with this new Hopper design, giving maximum contact between the disc and the Skittles:

  Additionally, to make the most of my new range, I also added some "shaking" action to the "fetching" part, hopping that it would always loose the Skittles:

  Although better, as can be seen in the video above, it is still jamming too many times and not allowing us to process an entire container, but I feel we made progress, so I will explore more in this direction...
  
  If you have any ideas please write it here :)
  
  That's all for now, see you next time!
  Cheers 

• The trickiest "simple" part of the project, the hopper design

  Gustavo Laureano • 01/28/2023 at 10:10 • 0 comments

  Since I started this project, my focus has always been on the design of the main mechanism, how it would work, what could go wrong, dimensions, jamming, etc., and in the end it went surprisingly well, I only needed a couple of iterations to adjust dimensions or clearances and it ran smoothly, as I show in my first video.
  

  Then the last missing part of the project, the hopper, which I thought should be something quick and simple, turned out to be very complicated, to paraphrase this nice HaD article:

  The first logical design was a simple funnel, silly me thought that skittles would play nice and just slide neatly through the pipe:
  
  

  
  I couldn't have been more wrong, as it turns out, candy coated in sticky sugar doesn't exactly have a low friction coefficient....
  
  So it was time for some research and iterations, first I thought that any minimal movement inside the funnel would loose the skittles, so I included a 'stirrer' made of paper clips in my design:
  
  

  
  But after only a few manual tests it proved itself completely useless, I would need a very large amount of displacement on the Z axis to have any effect, several times more than the grooves on the disc can provide, and even then it was still sometimes not enough
  
  Seeing that this task is actually more complicated than imagined I decided to search for solutions, and I even found a post on Reddit from someone with the same problem, following the suggestions there, I designed a new funnel with the largest possible bottom opening, in the hope that the disc movement and friction would cause some stirring and loose the skittles:
  
  
  

  And again, no luck, although better than before, it is nowhere near capable of avoiding jamming...
  
  
  

  
  So, after all these iterations, what's next?
  
  Everything points in the direction of a mechanical stirrer inside the hopper, something like those used in paintball hoppers:
  
  

  
  But I intend to stick to my original design, so it wouldn't make sense to use one more motor just for stirring, so my next step will be to design a stirrer connected to the main disk.

  
  That's it, see you next update
  

View all 2 project logs