casters

[gfish]

The one big question mark in the Kalamazoo's design is the front wheels. They need to caster, to allow for steering. And they need to be big and pneumatic, to smooth the ride. I did find some references to 25 inch industrial models that fit these requirements -- at a cool $1k each. So, uh, no. Going to have to make them myself from bike wheels.

Have any of you, my kind and loyal readers, ever done something like this? I can't find any guidelines on the web on the best way to approach this. There is a certain amount of caster effect in a normally-mounted bicycle wheel, but I have no idea if it will be enough, even if I increase the head angle. This would have the advantage of letting me use a standard bike fork and head tube pivot. (Though how I can cleanly and solidly mount it to the Kalamazoo frame is a bit of an open question.) Do people think that would work as a caster for a platform like this?

The other main alternative I see is to fab a new wheel frame with a pivoting swivel plate set back. Turn it into a standard caster design, basically. I'd rather not do this if I can avoid it, as there will be a lot of weird forces at work on this thing. And I'd still have the offset of the swivel plate as a tweakable parameter what would need optimizing.

These issues will have to be solved fairly quickly once I start construction, as getting the wheels mounted will be the first thing done after building the frame itself. There are too many questions (like how how close to the rear axle can I stand when piloting solo without risking flipping it) which impact later design choices that need to be answered empirically.

Comments

[technoshaman.livejournal.com]

The bit about being close to the rear axle can actually be answered with basic statics. What you really want is something that's slightly overstable in that direction... i.e. your own center of gravity should be forward of the rear axle, and not be counting on the weight of the front of the vehicle, particularly since acceleration is going to tend to lift the front. Basically you want the wheels as close to the edges of the vehicle as possible. (Indeed, I seem to recall some vehicles having the axle and the front/rear rail of the frame colocated.)

Two-wheeled and four-wheeled vehicles' caster setups are *entirely* different. A two-wheeled vehicle is designed with the steering axis *trailing* the axle, for dynamic stability. (The physics is weird, but understandable; I can loan you a book if you're interested, but it's kinda outside the scope here.) An actual *caster*, if you look at it on something like a wheeled tool box, has the steering axis *ahead* of the axle, so that the wheel is basically pulled along. This is what you want if you're going to steer with differential braking. You might have to fabricate this yourself; I'm not sure simply turning bike forks around backwards will work right. (This, of course, could be determined empirically... and possibly without removing the fork from the guinea pig bike :) But, yes, you'll want the head angle pretty much vertical, and the axle behind the steering head, regardless of how you do it.

Hope this helps...

(gee, I never knew learning to ride a motorcycle could help do Mad Science. :)

[gfish.livejournal.com]

Yes, obviously I want the CG as far inside the axles as possible. But as I don't know what the thing will mass, much less the distribution of the mass, it's going to have to be determine empirically. Worse, there are a lot of fuzzy ergonomic issues at play here. Just how close to rear edge is it comfortable to stand? How much do I shift during a pump cycle under load? What is the mount/dismount procedure like -- particularly the first, as I very well might need to give the thing a push to get the pump running from a dead stop, then jump on. Just too many issues. I'll make the frame as long as I can and still allow the parts to fit inside my car for transport, then scale down the pump arm until it works. I have a lot of gearing leeway on the sprockets in the drive connection, so I can always gear it farther down if needed.

I'm not sure you're right about wheel caster angle only helping for tilt steering. The steering axis of the wheel will still be contacting the ground plane ahead of the point of contact. Deviations from the vertical would thus be pushed back, even if the frame itself is help rigidly.

[technoshaman.livejournal.com]

Point. If you had the steering axis tilted as it normally is on a bike (i.e. so that the axis extended touches the ground ahead of the vertical plane of the steering head) but the axle aft of the steering axis, not only would you get trail stability, but turning the wheel would raise the front of the vehicle, thus giving you gravity stability (the wheels would *really* tend to center).

Wonder if you could make this adjustable easily?