Linear Path Shooter vs. Circular Path Shooter?

[Ether]

Originally Posted by Tem1514 Mentor

A multi ribbed flat belt will run at 3600 rpm with no problem at all on a 4 inch diameter wheel which translate to a very fast linear speed.

It translates into approx 21 mph, assuming no slipping.

3600 rpm / 60 min/sec = 60 rev/sec

60 rev/sec * pi*4/12 ft/rev = 62.8 ft/sec

62.8 ft/sec / 2 = 31.4 ft/sec

31.4 ft/sec * 3600 sec/hour / 5280 ft/mile = 21.4 mph

[Jon Stratis]

In Lunacy we ran timing belts instead of wheels in our shooter, in an attempt to compensate for the holes in the orbit balls. The belts were driven with only a small gear reduction from a CIM motor, and we never, at any point in the season, had an issue with them. They never came off, never wandered, no issues at all. A lot of design and work went into building a stable, adjustable system that would allow us to ensure everything was perfectly parallel, and using belts with teeth in them also helped.

I'm not saying I recommend going this path, but it certainly is possible to get your belts up to speed and keep them there without any issues.

[Tem1514 Mentor]

Originally Posted by

31.4 ft/sec * 3600 sec/hour / 5280 ft/mile = 21.4 mph

[I

[/i]

Ether. That 31.4 ft/sec sounds like it might do the job. I wonder what a CIM motor at full speed with a direct drive 6 inch wheel would be? Just don't have the CIM specs handy.

[MyNameIsPaul]

A circular shooter gives the Frisbee spin and more accuracy

[DjScribbles]

Originally Posted by Robogineer1649

What i can ask you is why would you use a half circle shooter that takes up much more space especially this year with smaller robots rather then building a linear shooter that works just as well?

It can be argued that 1 circular shooter may use less space than 2 parallel shooters, if you are trying to minimize the length of the shooter, and width doesn't matter.

[Ether]

Originally Posted by Tem1514 Mentor

Ether. That 31.4 ft/sec sounds like it might do the job. I wonder what a CIM motor at full speed with a direct drive 6 inch wheel would be? Just don't have the CIM specs handy.

CIM free speed is 5310 rpm. But you don't want to drive your shooter wheel at free speed. You need some headroom to accelerate back to the setpoint after a frisbee passes through.

[Ether]

Originally Posted by MyNameIsPaul

A circular shooter gives the Frisbee spin

so does a linear shooter

... and more accuracy

What is your source for this statement?

[billbo911]

You can see from some of our testing video (Thanks Kevin!) that we have tried a couple of different approaches. Granted, some of the tests aren't as valid because the mechanical setup's were a bit shoddy.
When you get to the 90 deg. circular shooter, you can notice some issues with the stability of the wall. We will be working on that tonight.
One really important take away from this testing is that increased RPM of the shooing wheel did not always translate to further distance. This is because the slipping that is occurring between the wheel and the Frisbee. Again, this is something we will be addressing tonight.

BTW, High Speed Video Capture (HSVC) is rapidly becoming one of our favorite new prototyping tools.

This is in Std. Res.
http://www.youtube.com/watch?v=p5bLi...ature=youtu.be

This one is in 1080.
http://www.youtube.com/watch?v=Wa6pc...ature=youtu.be

[ayeckley]

Some of the confusion/disagreement/misunderstanding/whatever seems to be because folks are unintentionally discussing different implementations of the linear shooter. The major sub-categories would seem to be "single drive side" vs. "double side drive" and "nip point contact" vs. "continuous contact". So far I've mostly seen videos of "single side drive, nip point contact". I haven't seen any videos of teams prototyping a "single drive side, continuous contact" design...yet.

[rcmolloy]

Originally Posted by skottmorris

Has anyone considered a linear path shooter for their design?

Everything I've seen on YouTube is built with a circular path. I'm curious if this was for design convenience or there is some magic to the disc following a circular path. My team has proposed creating a linear path shooter with two parallel motors to "spin" the disc down the launch path. Has anyone tried this? If it works, it would make a nice compact shooter.

Why not just quickly test both? You're going to have some discrepancies between the performance of the two anyway. Also, you might be using different materials than other teams and the such. Grab a bunch of wheels and have at it. You guys might also develop another idea just from doing that!

Good luck!

[CalTran]

I stand corrected on the belting. I misunderstood what was meant by belting; I was imagining that the belting in question was polycord or some other cord variant.

Originally Posted by ayeckley

Some of the confusion/disagreement/misunderstanding/whatever seems to be because folks are unintentionally discussing different implementations of the linear shooter. The major sub-categories would seem to be "single drive side" vs. "double side drive" and "nip point contact" vs. "continuous contact". So far I've mostly seen videos of "single side drive, nip point contact". I haven't seen any videos of teams prototyping a "single drive side, continuous contact" design...yet.

"NIP point contact"? Can you elaborate on to what this means?

[Siri]

Originally Posted by DjScribbles

It can be argued that 1 circular shooter may use less space than 2 parallel shooters, if you are trying to minimize the length of the shooter, and width doesn't matter.

This depends largely on wheel diameter. If you can achieve the linear velocity and contact time you're seeking using wheel diameter < the disc, even a dual linear can be shorter (final shot axis) than a 90deg turn.

[RSaunders]

One consideration for the circular vs linear shooter is how the Frisbee is released. The circular shooters I've seen in videos have a very gentle release, the outer wall very gradually moves away from the wheel until the gap is larger than the Frisbee. This means that small variations in Frisbee geometry, perhaps from prior rough handling, give slightly different release points. The theory seems to be that the Frisbee is "up to speed" well before the release point and so the exact point where it loses wheel contact doesn't matter.

The linear shooter used by the robotin3days folks uses a compliant wheel to hit the Frisbee with a pulse of energy. Given the speed involved, it it a very quick, violent, pulse. Ignoring the risk of Frisbee damage, there is no way to do that without adding some flex to the Frisbee and corresponding random velocity. I'm not at all worried that belts can't take the speed or forces required, just Google "serpentine belt" and see how they are used in cars.

Replacing the two wheels with two wheels connected by a belt, or 5 wheels and a belt if you want to go all-tank style, could mean you don't have to pinch the Frisbee as hard because you have more distance (= time) to add energy to it. It sounds like this could be better, but convincing experimental measurement of the magnitude of the effect is what we're lacking in this discussion.

[Jeffy]

Originally Posted by CalTran

"NIP point contact"? Can you elaborate on to what this means?

My understanding of this is that he means the very short contact time and small contact area that a wheel and a linear wall creates compared to a curved wall and a wheel.

[AdamHeard]

When making prototypes, keep in mind there are more variables than just linear versus curved.

Our linear prototype currently worked AWESOME, and our curved is merely pretty darn good.

We're smart enough to realize we haven't proved anything yet, and need to explore more.