Team 1678's Autocentering Roller
 

Here is a quick clip of our roller for this years competition.

http://www.youtube.com/watch?v=4QsjmuY_CP0

How do we do it? You guys are smart, I'm sure you can figure it out pretty quick. Neat huh? :D

Re: Team 1678's Autocentering Roller
 

Is that spiraled surgical tubing?

It's really effective

Re: Team 1678's Autocentering Roller
 

Great job!

How well does it work when turning quickly? We found that most types of rollers we prototyped similar to this couldn't effectively keep the ball from flying off when turning quickly (turning slowly was a different story). Were you able to accommodate for fast turning?

Re: Team 1678's Autocentering Roller
 

what are you using on the roller?

Re: Team 1678's Autocentering Roller
 

As BradMello said, it's probably spired surgical tubing. We're using spiraled rollers made of a different substance, and they too automatically direct the ball towards the center.

Re: Team 1678's Autocentering Roller
 

I thought that the robot can not hold on to the ball.

Re: Team 1678's Autocentering Roller
 

The robot cannot carry the ball. Carrying is defined as possession not in contact with the ground.

Re: Team 1678's Autocentering Roller
 

Thanks! To be honest we haven't tested the system a lot while moving, it retains the ball really well when turning and backing up at not-so-intense speeds, but thats not competition level speeds IMO. We'll test more, and in addition to fine tuning the height of the roller, we're only moving the drivetrain forward when auto-tracking the goal (aka, to turn left, move the right side forward instead of right forward and left back). This should improve our odds of keeping the ball, if only slightly.

The roller is made out of a solid aluminum 3/8" shaft, wrapped with a 1" wide strip of Wedgetop tread, fastened to the shaft with pop rivets. The right half spirals counter clockwise towards the middle and the left half spirals clockwise to wards the middle, which causes the ball to funnel towards the middle.

Re: Team 1678's Autocentering Roller
 

We are using a 3" PVC pipe covered with a loosely wrapped anti slip mat.
The pipe and mat came from Home Depot.
A CIM motor drives it at about 600-800 RPM.

Re: Team 1678's Autocentering Roller
 

Very nice! We're playing around with something similar ... still working on selecting the perfect material and/or spinner diameter.

One thing I did notice from the video is that the apparent point of contact is almost halfway up the ball. To me, it looked like it could've been more than 3" into the frame perimeter of the 'bot.

Make sure you take <G46> into account ...

Re: Team 1678's Autocentering Roller
 

Nice, but don't you think other robots can steal your ball?

Re: Team 1678's Autocentering Roller
 

I don't see how that could be possible due to this years rules.

Unless they lose control of the ball themselves, no one should be able to take it from them.

Re: Team 1678's Autocentering Roller
 

Looks great.

I have one question though. How do you plan on not possessing more than 1 ball at a time? It looks like that roller will possess up to 3 balls.

Re: Team 1678's Autocentering Roller
 

How do you figure the rules prohibit robots from stealing balls from other robots?

Re: Team 1678's Autocentering Roller
 

hey cans you post a picture of your roller stopped because were trying the same thing and its not centering for us

Re: Team 1678's Autocentering Roller
 

It appears to be almost halfway up the ball because we maximized our contact point. By using a small shaft and small pully, with bearings and brackets that are minimized (litterally some of the aluminum retaining the bearings are as think as the flange on the bearing :p) we ensure a high contact point on the ball. Our bar in the back of the roller cavity is at exactly 3" from our frame perimeter, so no possibility of having a ball come in any further than that. We make it a rule that the whole team reads the manual, not just the mentors on CD ;)

As far as possessing more than one ball is concerned, I don't see there being so many balls on the field that this will be an issue, and a mechanism capable of doing so isn't illegal in and of itself. Additionally, since the ball auto centers, it will be difficult for a second ball to fit in on either side.

I think I see where gorrilla is coming from. If the ball is 3" into your frame perimeter, and the bumper zone is 3 1/4-3 1/2" for each robot, thats makes for a distance of 1/2-1" between the outer edge of the ball and your opponents frame perimeter. Of course, this assumes bumpers don't squish, which is a poor assumption to make :rolleyes:

Re: Team 1678's Autocentering Roller
 

what is the spacing for your wedgetop

Re: Team 1678's Autocentering Roller
 

Because the rules say the ball can be held up to 3" inside your frame perimeter, and your bumpers will extend 3.25" beyond your frame perimeter. That means the soccer ball is 6.25" inside your robot.

That leaves only 2.75" inches of the ball sticking out beyond your frame.

An opponent robot trying to "steal" the ball also has bumpers 3.25" thick.

Even if the "stealing" robot drives right up against you, bumper to bumper, any mechanism they use to steal must then extend past the frame perimeter, but cannot extend past their bumper.

That means the stealing mechanism must also be able to retract within two seconds.

That also means that they only have 2.75" of the ball the interact with to try and steal.

It's not that it is impossible, but it is pretty difficult justify, especially over just concentrating on being able to kick well.

Re: Team 1678's Autocentering Roller
 

Please give us a stationary picture of the roller.
Bruce