1114 2009 Roller System

[jblay]

Like I'm sure many other teams are, we are trying to figure out a good way to rapidly feed balls to a shooter while avoiding any jams that seem like they will be common with these wiffle balls.
The one bot from the past that sticks out to me in ability to feed balls to a shooter quickly is 1114 from 2009. The thing is, I couldn't exactly figure out quite how it worked in 2009 and I still can't today. I read through their entire release thread found here and still don't have a great idea.
Can anyone shed some light on how their roller system managed to funnel those balls so no nicely?

[mman1506]

This is my outside opinion but I've seen similar systems on other robots (254 2012, 2056 2012, 341 2012 etc). The big difficulty with jams when handling a large amount of balls is usually due to indexing, the process of taking an unknown amount and position of game pieces into a controlled known amount and position. An obvious example of this is a funnel. Something that many top tier try to avoid is funneling/indexing in their ground intake before they have taken possession as it can often lead to jams due to forcing multiple objects into one space in an uncontrolled changing enviroment (the floor).

If you look at 1114 2009 or 2056 2012 you can see that their floor intake preforms little to no indexing. The balls can enter the robot at any spot along the width of the intake. They then enter a hopper that very loosely controls the balls. The position of the balls at this point is unknown but they are in possession by the robot. The balls then go into the indexing mechanism/funnel that takes the balls and puts them into a known position in the shooting tower. You can see in 1114's bot that they only place traction material in the center of the rollers leading to the tower and try to funnel the balls using gravity towards these center rollers rather than squeezing them into one spot. This greatly reduces the chance of jams as it is a constant and easily controlled enviroment.