FRC 2002 Very Fast Ball Collector

A few years ago I saw a video of a team that could pick up virtually all of the yellow balls along the wall in about 5 seconds. I believe the game was the 2002 FIRST competition. I was looking to see if anyone had the video of this. Does anyone know which team this was?

I think you may be referring to 2002, and if so there were a number of teams.

The team you probably saw was Team 95, who had an amazingly good ball collecting robot.

i seem to remember team 111 (wildstang) having a very efficent ball collector along with a crab drive that year, and also team 45 (tecknoKats) had a fast ball collector and a efficent storage apperatus

There is a video out there somewhere. I used to have it, but can’t locate it on this computer. If it becomes available to me, I’ll let you know. It’s quite short, and a little lack luster (the balls actually end up jamming before getting a full line so we only got about 15 of the 20 balls or so. We typically did better.)

The team still has a website at . Under the gallery look at the pictures from 2002, especially the construction shots. I was part of the build team for that 'bot (before joining 885 during college). If you have any questions, I can try to answer them as best as possible.

-Andy A.

Edit- I found the video on the teams website, hidden away in the sponsers section oddely enough. Here’s the link.

Here are bunch of ball collectors from 2000… Good luck…

team 111 (wildstang) and many pictures of their robot from 2002

Andy, do you recall how the rollers were powered? That machine is a monster.

Each roller set (two) were powered by a FP motor.

The motors ran one roller with a pinion and spur gear. I can’t recall the ratio. The rest in each set where connected by timing belts with 1:1 gearing so they all ran at an even speed. In theory each roller set ran at the same speed, but I can’t remember if we kept each motor spinning in the same direction to avoid directional bias.

The actual RPM range was adjustable by the driver, which proved to be a real art, as the power needed to displace the balls in the hopper went up rapidly as it filled. If you went to high early on the balls would just fly out of the hopper. To slow and they jammed up (this happens in the video).

I stand by the FP motors as an ideal (for FIRST purposes) for roller systems.

-Andy A.

what did the Wildstangs use for the conveyor belt(s)?

Elgin SHOULD still have the video from when he & I went to the Quincy Scrimmage - whether he still does is a whole other issue. :wink:

The 2002 Grasshopper robot from team 95 is the best ball collector ever in FIRST, imho. Simply the sound of that machine gobbling up 20 balls in about 4-5 seconds was astounding.

For those of you musically inclined, say the word “pop” 20 times in 5 seconds, by using 4 - 16th notes per second. Once you say it out loud and imagine a FIRST robot that could do this… it is pretty incredible.

Andy B.

Here are a few good pictures that should help you out

The accumulator
The conveyer
The shooter

But this design would beg a line of balls, right? It appears to me that once the balls lose their upward mobility and gravity sets in, the gobbler won’t pick any more up - it just pushes the balls in front.
This is an ongoing discussion in our team - an assembly that collects balls from the bottom of the hopper vs. one that deposits them at the top. We’ve reached a bit of an impasse…

You’ve got it mostly right.

The bot was designed with an opening move of grabbing all 20 balls in a line. Still, the rollers were the full width of the front, and could easily pick up balls randomly strewn across the field by simply driving into them. Regardless of where on the rollers a ball touched, it was sucked in. As the season went on, teams wised up to our strategy, and there were some good games of chicken as we ran down a line, and another bot tried to mess up the line to keep us from getting a clean 20. Being able to pick up the balls when they were not organized wasn’t as fast, but we still managed to get at least 20 balls every match, usually more like 30-40. The great width of the rollers was a driver aid, since aiming the full width onto a moving target is easier then a much smaller width.

As to gravity, this was a minor force in jamming as compared to the balls simply getting wedged in the hopper or the force to pass through the check valve. Even the slick soccer balls could bunch up and do odd things in the hopper. I expect the nerf balls to be worse. To counter act those forces, gravity, the check valve and misbehaving balls, the roller speed was adjustable by the operators with a pot but required a very deft touch to keep up with the influx of balls. As long as the rollers where up to speed, balls would enter with enough force to easily kick balls out of a full hopper. It wasn’t that the hopper filled up before gravity and jams overcame the motor power available. What happened to jam the rollers was that their speed was not increased quickly enough, and slowed down below a threshold. The gearing on the motors was such that they did not have enough torque to force the ball into a full hopper if the ball was already halfway in the rollers.

So the rollers relied on shear speed to overcome jams. This allowed us to gobble a line much faster then if the rollers had been running slower. A slower speed would have allowed more torque to clear jams, but might have caused more jams in the first place and would have required a lower bot speed. The speed down a line was important to this design, as we thought that we would be racing other teams for a line of balls. Turned out that few other teams did, and we kept going full speed to show off. Later on, opponents caught on and would try to mess up the line, so it did turn into a drag race.

Most of the time it worked out. As the season progressed the operator got better at it, which helped. What worked against us was the FP motors getting toasted, which lead to a constant drop off in available power. By the end of the season, they were pretty much shot. Replacements would have probably helped a lot in UTC.

-Andy A.