"Screw" Drive for FRC

[apples000]

So, for a project related to school, but no the robot team, I'm working on a drive system concept. The idea is to be able to have some type of new or unique omnidirectional style of drive, so I figured I might as well look into applications for FRC. Our group is planning on doing something like this(but much, much smaller ).

http://www.youtube.com/watch?v=1uynmApjhWI

The concept is kind of similar to meccanum wheels, but I don't envision it being able to turn the way swerve/meccanum do, as slowing down one side not only creates rotation, but translation to the side.

The design I'm working with is very simple; there's no CAD, and it has no need to be fast or pushy. It'll likely be powered with one 775 per side. Again, it's for a small robot that doesn't need to push. I think a really simple aluminum cylinder will work really well for the little robot, but I'd like to consider it for FRC too.

My question is, does anybody have any advice, recommendation, or ideas for this thing? Has anybody thought of/tried this before? Would it even be feasible for an FRC robot? The proof of concept definitely exists, but I have no idea what material I should make the thing out of.

[Henrique Schmit]

I believe that this could be done with mecanum wheels, you just need two axes with lines of cocentric mecanum wheels, but that would be extremely expensive. While this is a cool idea i can only see a small increase in pushing power happening at the cost of a good amount of design time and maybe a lot of money, so i don't think this would be very appropriate for FRC uses. If you want a render of my idea on how to do this i can make one.

[Chris is me]

If I remember correctly, the screw drive relies on driving on a deformable, soft surface to really move well.

Even if you never intend to push at all, I would never use less power than a single CIM per side in a robot that might play an FRC match someday. Pushing is far from the only reason teams put so much power into their drivetrains.

What potential advantages do you see over a more traditional mecanum drive?

[Cory]

Originally Posted by apples000

My question is, does anybody have any advice, recommendation, or ideas for this thing? Has anybody thought of/tried this before? Would it even be feasible for an FRC robot? The proof of concept definitely exists, but I have no idea what material I should make the thing out of.

What advantage over a conventional drivetrain would this give you?

If there is an advantage, why hasn't someone else already done it?

Two important questions to think about when evaluating a new design.

[apples000]

Again, I'm not saying this is a design that's right for FRC, I'm just throwing it out there because I'm doing it for a mini demo robot. The single 775 design would never be used for FRC. This robot fits in a cubic foot box. My friend came up with the idea, and I was a little skeptical that it would work for the mini robot, so I searched and found this video, which I thought was cool.

As shown in the video, it is able to go sideways, but I believe that the gearboxes used in the real videos do not allow rotation like that, so it is only shown rolling sideways down the hill.

To answer Cory's questions, I think the advantage would be a drive that allows the robot to slide to the sides with (possibly) more traction than mecanum, but I'm not sure how much traction this would actually have. It's also simpler in the fact that it uses fewer motors/fewer gearboxes.

Teams probably haven't done this because there is a good chance it will not work very well, so they didn't take the risk. It is not extremely easy to make the parts, and I doubt any team could get the design right on the first try.

As for the surface, it does appear to work on solid ground in this video, so it may not need a deformable surface.

http://www.youtube.com/watch?v=3uftJ3JJQk4

[Tyler2517]

I remember seeing a toy rc car that used this type or population when i was really little.


Research yaaa.
https://www.youtube.com/watch?v=JfYAFTCkkd0

I could see this being a strong FTC drive train when there is a large amount of small objects on the field.
You could also put the motors for movement in the screws. I don't think the tilt feature would really be needed.

[caboosev11]

I have a feeling this would be a little rough a on carpets.

[apples000]

Originally Posted by caboosev11

I have a feeling this would be a little rough a on carpets.

Why? If the thread were to be covered in tread, I believe it would work well. You would have a thinner contact area than a typical wheel, but you would contact the ground in multiple places.

[scca229]

Originally Posted by apples000

Why? If the thread were to be covered in tread, I believe it would work well. You would have a thinner contact area than a typical wheel, but you would contact the ground in multiple places.

If I'm not mistaken, doesn't a screw drive like this use the "slipping" of the screw thread along its length to achieve the forward momentum of the vehicle? Sure sideways would benefit with having tread along the edge of the thread (say that 5 times fast), but forward/backward would then suffer.

Am I right here? I've been Youtube walking through the videos showing this type of drive and pretty confident in my impression.

[asid61]

Originally Posted by Tyler2517

I remember seeing a toy rc car that used this type or population when i was really little.


Research yaaa.
https://www.youtube.com/watch?v=JfYAFTCkkd0

I could see this being a strong FTC drive train when there is a large amount of small objects on the field.
You could also put the motors for movement in the screws. I don't think the tilt feature would really be needed.

HOLY COW I never even knew that existed.
I really hope this drive train gets made. It seems to have a few advantages, such as:
1. Small footprint. 4" on each side of the robot or even underneath it.
2. Only 2 gearboxes needed.
3. Omnidirectional, although you would want to test pushing power. I think it would get very good traction in the forward direction.

If it turns out to be more than a gimmick that would be cool.
McMaster sells some conveyor screws that might be suitable for prototyping, but they are expensive. You could put tread on the them with some modifications.

EDIT: Some creative lathe work could do it too.

[IndySam]

Always hopped they would bring the corn field back so we could try this out.

[Foster]

Originally Posted by IndySam

Always hopped they would bring the corn field back so we could try this out.

Is it too early to start a "2015 The Marsh Game" thread?

I think the damage to the carpet would get you asked to leave.

[Mr. B]

Maybe not in FIRST but, it has beem done--My son had one!

http://www.sndkids.com/hot-wheels-r-...&utm_campaign=