Re: Linkage Design Problem
 

Andrew,
It would seem that frictional and side forces would diminish in the design, but in fact I think they are just transferred to a different part of the mechanism. Hopefully a mechanical person will step in here. Think about your scenario, the force of the robot pushing you is transferred to the floor through the pivot point for your steering assy. and wheel, bearings, etc. However, a move to disengage will add to those forces the turning forces that occur when the linkage is actuated to rotate everything. That plus the added weight of the robot puts a lot of friction on that vertical shaft about which the whole assembly is turning. Now that we have a drawing of your proposal, run some rough calculations on the amount of space you need within your robot for the mechanism and I think you will see why crab systems become more a vertical assy. If you can commit to the real estate, this might prove an interesting design.

Re: Linkage Design Problem
 

You're completely right about the space requirements Al, and they'll run easily a square foot a module. However, we've used a six wheel drive platform for quite a while now, and that takes up even more space! I'll have to definently design something first, and see what type of resources and space that'll burn up, but I think it's a feasible spin on swerve drive, under the correct circumstances.

Re: Linkage Design Problem
 

How did you come upon the requirement of one square foot per module? That seems very, very large for what you're trying to implement unless you're including the full length of the linkage -- but even then, it seems as if it could be quite a bit smaller.

Re: Linkage Design Problem
 

Are you planning on four of these units per robot? If so, are the pivots all moving in the same direction or is their a mirror layout with two pivoting one way and two the other?

It looks like you will have to drive the wheel at the same time as you pivot in order to avoid the turning friction of the vertical crab design. There will most likely be some frame motion during the pivot due to differences in the speeds and programming. The wheels may be scrubbing as they move out of sync with each other.

This is an interesting design. Take a moment to evaluate the interaction between all wheels with the robot frame and the floor. Also do an evaluation on pro/con of effectiveness if you plan on using on your FIRST robot next year.

Re: Linkage Design Problem
 

I think this could be a problem. To swing all four wheels in an arc... how does that work? does the robot pivot on the wheels as they swing? That means the robot has to move to change directions. If you are up against a wall then the wheels would have to be powered to switch.

Also, the piston now needs to hold its position. When the wheel is exherting full force (pushing another robot...) the wheel will want to pivot to the other position. The only thing stopping it would be the piston force.

But...this could be a good thing? Maybe you dont need the pistons at all. Maybe all you need is a latch in both positions. To turn the wheels you release the latches, and back up sharply (or move forward sharply) - the traction of the wheels will twist them around to the other position, where they lock in place.

Maybe?

Re: Linkage Design Problem
 

Heres another terrible Paint drawing that may or may not explain things better. There are four independent modules, with pivots in the center. All modules must move generally together, although it should be so fast that it won't make a difference. Turning the wheels, or making sure that they're recieving no power during a shift will allow it to move smoothly. Pnuematics should more or less be fast enough and forceful enough that syncronization won't be a problem.

M. Krass, that area figure was just a guess, taking into account the framing around the module, a beefy support frame, and the area required to arc a 6" wheel.

Ken, you are completely right about piston force needing to keep the wheel in position. I really haven't put too much thought into this, and the original idea was that a complex linkage would be used, and it's structure and friction would more or less soak up some of this force. Not so anymore, so perhaps I'll need to think up a locking mechanism that engages when the piston fully moves the module.

Re: Linkage Design Problem
 

I'm liking your general idea, very innovative. However, 6" wheel seems quite large for this application. You may want to consider making your wheel smaller (3"?), that would decrease your module size. If you bring down the wheel size, you can also run your pivot point right on top of the wheel, which throws out the whole frame swinging issue, or wheel under power while shifting. This idea has really made me think. From this idea, you could make quite an interesting bot. Good work, and I look forward to seeing your progress on this idea!

Re: Linkage Design Problem
 

This brings into play an interesting problem in forces. As the wheels attempt to change orientation their rotation is counter to the robot. So do the wheels move and the robot stays in the same position, do the wheels stay and the robot moves? Could have some interesting action in competition but will it give you an advantage? Don't forget you have to include the arc of the wheel diamter in your calculations. In your drawing, an assembly rotating to a new position, inscribes and arc with both the leading and trailing edge of each wheel. The clearance on a 6" wheel, off the top of my head, will be about 1.5-2 inches greater than the length of the assy. from pivot point to outside structural member. With that kind of clearance, you might be able to gain an a little in the outside dimension of the outer axle support.

Re: Linkage Design Problem
 

Well, the concept could be completely adapted for a three inch wheel, but we have and will continue to use our six inch elevators- they're just too grippy! Short of building a new rubber mold, we've just got the sixers. But again, your completely right, and three inch wheels would even make the gearing easier. As far as moving the pivot point, you could, but I'm looking to build a stronger pivot all in all. With a three inch wheel, an over center ivot would reduce forces, so it would be more feasible.

Re: Linkage Design Problem
 

This design of yours keeps perculating in the back of my head, and I keep thinking there has to be a way to do this

if you put the pivot point behind the wheel, instead of next to it, then the pushing and pulling force of the drive-action of the wheel will not have any tendancy to turn the wheel to the side.

like this (top view of wheel and pivot point)

You might also design this so that one position pivots the wheels outside the normal frame of the robot, giving you a much longer wheel base (more stablility) - the other postion pulls them into the starting-outline of the frame, pointing at a right angle.

Re: Linkage Design Problem
 

Right, and this is what I had originally. However, this would mean that you have to drag a highly tractive wheel sideways, instead of rolling it. But I'm glad it's got you thinking! I'm thinking about a vise grip type locker at the ends of the arc. But the problem is disengaging them when you want to move.

Re: Linkage Design Problem
 

Why is it that you don't want to pivot the wheel about its center?

Re: Linkage Design Problem
 

Having built industrial assembly machines for many years, this is the issue that I would be most concerned about. When a machine is idle, the storage tanks and supply lines fill and achieve full pressure. When valves open the air expands into an area of low pressure causing a drop in pressure. The more valves, the more drop. The more drop, the slower the response. We counteract this by adding more storage, larger lines and dedicated lines with individual connections on the tank.

On our robot we see this issue when we need to move cylinders multiple times and need air faster than the pump can resupply. Add in the restrictions on storage tanks and extraneous tubing and you can see why I would be concerned.

I suggest that you build a quick prototype to test your hypothesis. The data you collect will provide direction on your decisions for wheel diameter, linkage length and force requirements. Don't underestimate the interaction of the floor or the movement of the frame in this mechanism.

Good luck!

Re: Linkage Design Problem
 

you dont really need to skid the wheel sideways, it would act like the castors on the front of a shopping cart - the wheel will pivot on the center of its contact with the floor. The wheel just needs to be able to roll to get to its new position.

This would be a good thing to mock up with a VEX or Edu kit to get a feel for the action.

The lock on Vise Grips is called an over-center latch. You can design them to work with very little release pressure. This is what they use on some ambulance gerneys - when they pull it out of the ambulance the legs drop down and latch. When they push the gerney against the rear bumper of the ambulance the over-center latch is released and the legs fold back up.

Re: Linkage Design Problem
 

Ken I was thinking about this last night, and I think I realized what you were getting at. Instead of moving the module via pnuematics, simply move the module by running the wheel. Use a cylinder or something else to lock it in each position. I really, really like this idea, but I'm worried that if a wheel was to pick up off the ground, it would not move and engage correctly. But it would allow you to use a single lockong cylinder for all four modules.