Double Shooter for 2013 robot
 

Team 701 is planning on making another 2013 robot for the Western Region Robotics Forum Cal Games competition in the fall. The main goal we have is to make a double shooter. I understand that team 2337 The EngiNERDs has a double shooter, and I'm not sure if there are other teams that have a similar designs. From the videos and pictures I saw, they have the shooters on top of each other and they are using pneumatic wheels. Our team is planning on having the shooters side by side using 4" polyurethane wheels, but the wheels will be on opposite sides, so it's just like a standard two wheel linear shooter, but with a second shooter next to it, just mirrored.

I would like to know if anybody else has tried a side by side double shooter, or any type of double shooter, and has had any success with it, or if it was a failure, I would like to know what was learned from that.

Re: Double Shooter for 2013 robot
 

Check out 1503 Spartonics. They were pretty successful with their slingshot shooter. Technically, they have a quadruple shooter. Probably not what you had in mind for a multi disc shooter, but it should fuel some more varied design discussion.

I was talking to a mentor from 1503 at Championships and he said that one way to make cycle times faster is to reduce the amount of time between disc shots. And the best way to do that is to shoot multiple discs at a time.

Re: Double Shooter for 2013 robot
 

Yeah I remember watching 1503 at championships and I was surprised that anyone actually tried that design, but we want to be able to do a 7 to 9 disk autonomous, and I'm not sure how simple a floor pickup and load for a multi-disk slingshot would be. We also want a below 30 inch robot, and a low slingshot would be quite unreliable, and at that angle, fitting four disks at the same time in the high goal would be difficult, if even possible.

We are pretty much locked in on wanting the side by side shooter at the moment. We have until September to start building, so there are definitely going to be some improvements or changes to the design. We're betting that the robot would have to be quite wide to accommodate the diameter of two shooter wheels, two disks, and the thickness of our metal stock all in a row across the width of the bot.

Re: Double Shooter for 2013 robot
 

Why did you decide on a double shooter? From a design perspective you would be much better off just making a single shooter that shoots twice as fast. Two shooters are twice as heavy and require twice the tuning and maintaince of one shooter. Even if you were just designing and building it for the experience you would still probably learn more attemping to optimize a single shooter.

, Bryan

Re: Double Shooter for 2013 robot
 

We proto-typed a double shooter and thought we could make it work.

In our release video at 0:28
http://www.youtube.com/watch?feature...MCYh1Prs#t=28s

You can see this being shot.

We found you have to delay the second shot, the disks could not fly ontop of each other 100 millisecond delay will work.

We moved on to build this double deck design as seen here in the build season pictures. You can see several pictures.

http://www.flickr.com/photos/teamthr...57632841870583

But later, we could not separate the frisbee stacks into two channels in this manner. So after fighting it for a week or so...we gave up, closed the second hole. But what we found was that the second wheel gave us the momentum to shoot rapid fire. The total weight of the shooter wheels are 2.2 lbs. This allowed for some rapid fire with no loss in speed.

Later we improved this to shoot around 1.3 seconds all 4 disks with a single shooter. As see below. 2-3 disk are in the air before the first one hits the goal.

http://www.youtube.com/watch?feature...&v=yI_XldiEXFM

That's kind of our design story. We learned lot from this.

Re: Double Shooter for 2013 robot
 

To clarify our design, each shooter uses a 8" AM pneumatic wheel with 90" turns. We prototyped linear shooters using 8" AM wheels and we couldn't get our hands on the BaneBot wheels (see 67, 3847's shooters) in time for development, but the original goal was to have two linear shooters side by side, as you are proposing.

Our first strategy this year was to focus on autonomous. The original strategy of the double shooter was to allow "shotgun" blasts from closer ranges, such as touching the alliance wall (like Rebound Rumble robots touching the fender shooting lay-up) and the pyramid. Shooting two at once would speed up our fire rate - especially during the 15 second period, we believed - and we thought it was within our capabilities (spoiler: it ended up not being). What we really should have done was watch match videos from 2006, and realize that it would be best to optimize firing speed, not double the output.

The main problems this posed with the rest of our design were feeding the shooters with discs - taking them from the ground, and queuing them for the shooter. Integration. If we had devoted design resources to slick integration between our ground pickup and a single shooter, we would be able to shoot a lot quicker. Look at some of the most successful teams this year: 33, 469, 254, 118 - all of their subsystems for moving discs from the ground or feeder station to queuing for their shooters were robust and very efficient.

We also struggled with having each shooter shoot at the same height (we were still playing with it at our fifth event this year, the Championship)

From our experience this build season, Bryan is fairly dead-on, and I give you the same recommendation he does. There has not been a demonstrated performance benefit (you can compare the performance of teams that had double shooters such as ourselves), and it essentially requires double the time/design resources for packaging the design, and mechanically tuning and maintaining each shooter individually. Weight was also an immense (pun intended) struggle for us this year, much more than other years.

I expect rebuilding your 2013 robot would be a lot more realistic and a lot more successful in competition with a single, efficient, quick shooter.

Re: Double Shooter for 2013 robot
 

I can see it now a 1503 style shooter mounted to a four bar linkage with a pickup at the end :P

Re: Double Shooter for 2013 robot
 

I would highly recommend spending the time to just make a shooter twice as fast, its probably a more worthwhile exercise. Building a shooter that can shoot 4 discs in ~1.5sec is a challenging exercise that will stretch your iteration skills and programming talent. Every year the thing that separates the best teams from the rest is how quickly they can score, learning how to continuously iterate on your scoring mechanism to make it faster/more effective is a very important skill for any FRC Team.

Between Waterloo and Champs our mechanical and programming teams spent weeks iterating on a new shooter for champs. Using two mini CIMs instead of originally one, we iterated on everything from different speed control algorithms, to hopper designs to make sure the discs stack and fall properly into the shooter, new mechanism to pull the discs into the shooter wheel, and many iterations on the loading trey. This dramatically reduced our shooting time at the pyramid, and served us well at Champs... :)

Re: Double Shooter for 2013 robot
 

In theory, a side-by-side double shooter isn't difficult to build- what makes it difficult- especially if you also want floor pickup- is successfully indexing the discs. In addition- you're going to need to design with the number ofmotors available in mind - at least one CIM Per shooter, 4 on the drive train, any sort of drop-down intake will make use of another, etc.

I'm interested to see how your new robot comes out and performs.
Keep us posted on your progress!

Re: Double Shooter for 2013 robot
 

There is little reason to use a CIM for each shooter when you have 4 mini CIMs. They are lighter, spin a bit faster, and have sufficient torque. Also, you can spin the shooters off of the same motors, and while your spin up time might suffer, you the time it takes to get all your frisbees out is 1/3 of normal. 3 mini CIMs/2 CIMs powering both shooters will probably be fine.