pic: New Drive System for 703

[Reaper40]

Won’t the wheels take more power?

[ZeetherKID77]

This is made of win.

[Ericgehrken]

Almost as much grip as tracks minus the breaking of belts. I think it's a new record for FIRST as in most drive wheels on a robot.

[jcatt]

Very Nice, Very Nice.

Cowmankoza-
very nice, i like the design, im assuming the bottom wheel is offset?, and do i spy geardrive?, my team was contemplating doing something similar, kudos to making it work

One thing that our mentors noted aboout this design and what in the end discuraged us from doing it is. That if one of your gears fails or binds then you loose all on your wheels. The only way to insure this doesn't happen is to design some fail safe into each gear.

At least with chain, if you chain each wheel indiviualy then if one chain breaks you still have the use of some of your wheels.

But still great work, We wil have to take a better look at the design, during the off season.

[groves]

If the center wheel is offset, then you have a really complex six-wheel drive. making 8 of the wheels either not touching the ground or not effectively getting traction. If they are getting traction then turning will become pretty tough depending on the amount of motors your using and then battery use comes into play. ignoring what I just said I really think it's a neat drive system.

[James1902]

This looks really cool and I really want to see it drive. But just some quick questions.
How many motors are you using to power either side?
Can it climb a ramp efficiently?
How easiley repaired is it if a gear or wheel breaks or becomes unalinged?
Otherwise SWEEEEEEEET

[Eric Scheuing]

Do you guys have a CAD team? I feel bad for whoever has to make that. Design Accelerator FTW.

[Lil' Lavery]

So many people have asked why, and I'll try and elaborate on what Billfred and the members of 703 have said.
This design effectively emulates many positive features of a treaded drive system, as well as eliminates some of the negatives of chain and belt driven drives. The close proximity of the wheels allows for the robot to have a very, very, low ground clearance and still have the ability to climb ramps, steps, and platforms without bottoming out or high centering. The greater quantity of wheels creates more contact area with the ground, which in turn allows for a more even distribution of weight (although, still not nearly as well as a tread in this respect). The fact that all 14 wheels are driven allows for any wheels to lose contact with the ground (such as when traveling up an incline) and for the robot to still have the ability to drive. The use of intermediate idler gears instead of chain or belting eliminates the risk of the chain/belting popping off or slipping. It also may have actually saved weight depending on the size of the sprocket/pulleys and the chain run (if they would have chosen a "staggered" chain run, so that a single chain failure doesn't result in the failure of the whole side of the drive, it probably would have weighed much more than the gears).
There are some cons to this drive system as well. Spur gears are slightly less efficient than chain (not by much though), but so many gears magnifies that slightly. Additionally, while failures should be far less often, if/when they do occur, it is likely to be far more spectacular and require probably a more complex fix than if they had used chain or belting (although, once again, this design was chosen to try and make sure failures wouldn't occur). Additionally, the 14 wheels themselves are far heavier than the 4 or 6 pulleys they would have likely used in a tread drive.

[cbale2000]

Originally Posted by Cowmankoza

Upon looking at the design again (I still love it), my only concern is the side panels, are they lexan? I'm just worried abotu a side hard impact shattering them and leaving gears all over the field. But good luck to you guys!

We do plan on using Bumpers this year (much to my dismay) I just hope they don't interfere with robot operations like I saw some did in last years game.

Originally Posted by Eric Scheuing

Do you guys have a CAD team? I feel bad for whoever has to make that. Design Accelerator FTW.

We don't really use CAD that much (We help design it but our mentors tend to handle making the blueprints of the parts). We've modeled our robots in Inventor before but it's usually just for the sake of doing it (though we have found it useful when trying to make an animation).

Originally Posted by Lil' Lavery

So many people have asked why, and I'll try and elaborate on what Billfred and the members of 703 have said.
This design effectively emulates many positive features of a treaded drive system, as well as eliminates some of the negatives of chain and belt driven drives. The close proximity of the wheels allows for the robot to have a very, very, low ground clearance and still have the ability to climb ramps, steps, and platforms without bottoming out or high centering. The greater quantity of wheels creates more contact area with the ground, which in turn allows for a more even distribution of weight (although, still not nearly as well as a tread in this respect). The fact that all 14 wheels are driven allows for any wheels to lose contact with the ground (such as when traveling up an incline) and for the robot to still have the ability to drive. The use of intermediate idler gears instead of chain or belting eliminates the risk of the chain/belting popping off or slipping. It also may have actually saved weight depending on the size of the sprocket/pulleys and the chain run (if you they would have chosen a "staggered" chain run, so that a single chain failure doesn't result in the failure of the whole side of the drive, it probably would have weighed much more than the gears).
There are some cons to this drive system as well. Spur gears are slightly less efficient than chain (not by much though), but so many gears magnifies that slightly. Additionally, while failures should be far less often, if/when they do occur, it is likely to be far more spectacular and require probably a more complex fix than if they had used chain or belting (although, once again, this design was chosen to try and make sure failures wouldn't occur). Additionally, the 14 wheels themselves are far heavier than the 4 or 6 pulleys they would have likely used in a tread drive.

You're right on there, couldn't have put it better myself.


I will just add though, we figure if a single wheel came off, we have each wheel independently connected to the chassis using easy to remove pins, so if something happened to one wheel, all we have to do is pop another one in. Now if something major did happen and we lost a whole side it probably wouldn't be any harder than replacing a tread as it would require nearly the same actions to do.
Every year we get progressively better at making the robot fast to repair, this year will probably see fastest repairs we've made yet.

[LWS]

A good bit of the discussion seems centered on how effective the design is in getting all of those wheels to work (e.g. no chains, etc). The real question it seems is, why have all of those wheels? The only effective answer I have seen so far is to avoid "high centering" when going up ramps.

The common response to all of those wheels is "they would give lots of traction". But will this approach be more effective in doing so than one with fewer wheels? The simple model of friction (which is what traction really is) states that maximum friction force is the coefficient of friction multiplied by the normal force (i.e. the weight pushing down on the wheels). The coefficient is a function of the material of the wheel and the carpet. Assuming that there is a given weight for the robot, this weight would be distributed over the wheels, and if there were more wheels, there is less weight per wheel.

The short story is that adding more wheels in this case does not add more traction. The fallacy that it would comes from the "outdoor scenario", where terrain may have low coefficients of friction (e.g. snow). The idea here is that more drive wheels can avoid the situation where wheels slip, and the friction coefficient lessens (going from static to kinetic friction).

Wikipedia has more info for those interested.
http://en.wikipedia.org/wiki/Friction

[EricH]

Originally Posted by LWS

A good bit of the discussion seems centered on how effective the design is in getting all of those wheels to work (e.g. no chains, etc). The real question it seems is, why have all of those wheels? The only effective answer I have seen so far is to avoid "high centering" when going up ramps.

The common response to all of those wheels is "they would give lots of traction". But will this approach be more effective in doing so than one with fewer wheels? The simple model of friction (which is what traction really is) states that maximum friction force is the coefficient of friction multiplied by the normal force (i.e. the weight pushing down on the wheels). The coefficient is a function of the material of the wheel and the carpet. Assuming that there is a given weight for the robot, this weight would be distributed over the wheels, and if there were more wheels, there is less weight per wheel.

The real reason is that the team wanted a treaded robot without the treads. (As in, the best points of both wheel and tread systems.) We shall see whether they succeeded or not when they compete.

[Lil' Lavery]

Originally Posted by LWS

A good bit of the discussion seems centered on how effective the design is in getting all of those wheels to work (e.g. no chains, etc). The real question it seems is, why have all of those wheels? The only effective answer I have seen so far is to avoid "high centering" when going up ramps.
[/url]

Don't forget lesser weight (and a more even distribution of weight) on each wheel. This means that less force passes through each wheel, meaning a lesser chance of failure (especially of items such as treads...or the carpet they touch).

[cbale2000]

Originally Posted by Po-ser

I think "Holy moly!!" about sums it up.

Do update and tells us how it drives once you hook it up. Also, once you figure out all of the numbers, could you tell me how much more it would weigh than your tread drive train?

That's really something.

I don't know the exact numbers but I believe someone mentioned it was a little under than twice the weight of our previous drive systems. I'll get the exact numbers as soon as I can.

Originally Posted by Alex Cormier

i am very impressed and that does not happen often. One way i have noticed to drastically reduce weight is to go lexan wheels. Get a big block of it and make the wheels, 1126 has been doing that for a few years and the base gets better and better every year.

We thought of something like that but we decided we didn't want to compromise the integrity of the wheels. We figured we could get away with drilling holes in the wheels and gears plus we plan to extensively use Carbon Fiber composites.

Originally Posted by Reaper40

Won’t the wheels take more power?

We suspected it probably would however our initial tests have shown us otherwise. It actually appears to take less effort than our previous tread drives!