Little girl's off-rail industrial hand car

[BoyScoutGirl]

So. My friends have invited me to work on a "little" project for the burn:


The hand car will be entirely human-powered, with a newly-fabricated frame but scrapped out parts, such as suspension.

I would love to hear your collective thoughts on this project, eplayans. Specifically, we're currently ruminating on:

[*]Steering mechanism. Rail cars didn't have to deal with this, obviously, so it's the first point to address: how do we include something that pivots and also bears the full weight? How do we use a pump handle to create a turn in wheels under us?

These guys turn by moving the front two pairs of wheels while keeping the rear pairs straight (I think):


So, we're considering using eight wheels. Which brings me to...

[*]Number of wheels. Four gives the classic image but we are not necessarily aiming for a railroad aesthetic. We just want the thing to ride well and make it through the week in one (preferably functioning) piece.

[*]Wheel size. We're currently looking at cannibalizing kids' bikes for 14" wheels, but will their tread be wide enough? Should we be thinking about something wider?

[*]Wheel position. Under or alongside the platform? Visually, we'd prefer under, and this will give us more space to work with given our fixed max width: We don't want it much wider than 4'6" so it will fit in our trailer. For reference, railroad tracks in the USA have about 4'9" between the rails, so the size seems about right. We'd love to fit four people, two on each side, but we'll see if/how that works out.

[*]Gearing. It's tempting to include shiftable gears to help us get through dust drifts, but that's also one more moving part that could be mucked up.

You guys have inspired me so much with your creations, so I'll be pleased to share with you our progress in this thread, should this project come to fruition!

[Ratty]

Boyscoutgirl THAT is an excellent idea you have working. I'm sure the bike and car builders around here can point in the right direction.

Maybe think a little beefier on the tire size. You will be adding a lot of weight with just 2 people. Maybe beast tires? I can't wait to see this.

[Captain Goddammit]

This has been done before, an eplayan named LeChat Noir built a pump car. There's an old thread documenting its construction, it was called The Contraption.
It ended up being too much a pain in the ass to ride and it ended up with a motor on it, but it was built much heavier.

Kids bikes have 16" wheels. Ratty brought up a good point about weight. Four people plus whatever this thing itself weighs may end up "taco-ing" those ultralight duty wheels.
Don't even consider the tandem wheel layout of that big military vehicle, you'll have way too much drag when you try to turn. It works on the army truck because it has a big Diesel engine and low gears. It can laugh off a good amount of drag, you cannot because your power-to-weight ratio will be so minimal.

There are a lot of stronger, wider wheels from things like riding lawn mowers and go-carts you might look at. If you use bicycle wheels, I think you may want to consider using Skyway Tuff Wheels. Those are made for the BMX bikes. They're pretty light, not terribly expensive, and strong as hell. I had a set on my Mongoose BMX in the '80s - I still have them - I beat the crap out of them including running them over with a pickup truck once. Zero damage. They still make 'em, and they're available in 16". Try PlanetBMX, good source for such stuff.

If I were building this I'd use three axles. Drive the middle axle with the pump, and have each outer axle steer. Clever linkage will be in order. Extra cool bonus points for doing that so it would be truly double ended, it would have no front or rear, it'd work exactly the same both ways. Bonus advantage, the center drive axle would be directly below the pump handle which will eliminate some complication and increase efficiency.

For steering there are two ways you can go: use a solid axle that pivots in the middle, like a Radio Flyer little red wagon, or have the wheels hinge at the ends of the axle, like cars do. If you do the car style, Google "Ackerman angle". You need to get that right.

I think the simplest way to accomplish the three axle layout would be to use just a single wheel at each end. One front wheel, two driven middle wheels (one left and one right) and one rear wheel.

I think I'd build it as a single speed first and try it out, then add gears if needed.

LeChat Noir's rig had a separate driver who did the steering. If you don't want that I'd suggest a tiller that one of the people can reach with one hand. To be REALLY fancy, set it up so someone on either end can steer, so you never have to look backwards when going the other way.

[Thecatman]

It'd be cool to have a real one on standard 4ft 81/2in gauge rail as shown in our first picture. It'd take alot of real estate, but it'd be neat to see something like that.

[BoyScoutGirl]

Thanks for the responses and enthusiasm!

Catman: while it would ride smoother and be awesome to actually have a rail in place, that is beyond the scope of this project. Honestly, looking at LeChat's past threads, this project may beyond the scope of this project! We're certainly behind on time.

Captain, I was hoping you would chime in - we're grateful for your feedback and ideas. Sometimes it just takes knowing the right vocabulary to look up - from "Ackerman angle" to "contraption!"

Researching BMX tires was on our to-do list already, and those Skyways have a really great look! That company offers some with white tires, which was exactly what my co-conspirators were looking for aesthetically.

With three axles as you describe, I think clever undercarriage lighting could make it look like the thing only has two wheels when seen from a distance. It'd be a laugh to convince people that we were perfectly balanced over the center axle.

[Tiahaar]

Hiya BoyScoutGirl, Captain Goddammit alerted me to your interesting idea and opined that I ought to add some of my crazy thoughts to the thread.

Love the handcar thing! As it happens there might be a solution to steering to be found under a skateboard...you could use tilting bogie carriers on your axles so that shifting your weight makes the thing turn.

You could get away without any chains as well, the axles could have a crank point in the center driven by rods connected to a central distributing crank that is in turn driven forward or backward by the handcar handles. Shucks I might have to make a small one to try this!

Would recommend only driving one wheel on each axle, each on the same side and letting the other side's wheels freewheel...otherwise you'll tend to bind up in turns.

Great project!

[BBadger]

Man, that sounds like some back-breaking effort. You might also consider some other hand-car-type locomotion mechanism than the standard Sheffield handcar design. There are some improved designs that might transfer power to the wheels more efficiently.

You could also change the primary method of transferring human power to the wheels. One problem with handcars is that you're using your arms and legs to actively power the wheels, rather than taking advantage of gravity to assist you for half a cycle.

Instead of lifting and pushing a lever with your hands (and hurting your back), what about a teetering platform that moves up and down so that you can jump-pump your legs up and down with gravity to power the cart. To keep yourself steady, you'd grip a horizontal rail in the middle or something. Since your arms will not be actively involved in powering the machine, you can then steer the machine more easily too.

Assuming you can make it safely, the mechanism could even be a teeter-totter type design. Then you can sit and pump your legs instead of standing all the time. Depending on size of the cart the motion might affect the stability though, so maybe the platform would work better.

Anyway, just throwing it out there.

[Tiahaar]

a quick paint drawing of what I was thinking of in my usual crappy style:

it is not very close to scale, possibly not buildable, and certainly improvable

[Ratty]

Bbadger! A teeter totter car is a great idea. I imagine a young couple gently rising up and down as they cruise the playa. (Of course this would require a hidden electric motor.) but it's a nice visual.

[BoyScoutGirl]

I LOVE the teeter-totter idea and I think we could totally swing it. With careful planning we could maybe even make it convertible: when the arms get tired, switched to seated, and vice versa. One of the co-conspirators was actually thinking of a similar idea when I read your response, BBadger.

Tiahaar, thanks for stopping by. I loved reading your input in LeChat's threads, so I'm glad the Captain sent you our way

We really like the idea of steering by leaning, as skateboards do. That would be really fun, and it would make the design elegant. However, I see some problems with this stemming from the fact that skateboards are freewheeling. How would we get the drive train to connect past the independent truck? Would it be difficult to power the wheels while turning? We don't want to have to build custom differentials...

Also:

Would recommend only driving one wheel on each axle, each on the same side and letting the other side's wheels freewheel...

Are you saying we should only paddle this canoe on the left side??

[ygmir]

wheels/tires and weight seem a choke point.
you'll have to design for the unexpected, as in too many people trying to hop on at once.
if you could find some old ATV tires, and maybe a drive axle (with sprocket) , they'd be soft for suspension, yet made to take a lot of abuse.

a simple tie rod steering system should be easy enough?

[tatonka]

[Captain Goddammit]

While I think the skateboard style lean steering idea is brilliant, I think with four people on it you'd have a really hard time maneuvering.
You have to either have a differential or just simply drive one wheel and let the other freewheel - which is how most adult size tricycles work - or it won't want to turn.
I've lost the info, maybe Tiahaar has it, there's a perfect little diff available from an industrial supply place. But really, you don't benefit much if any from using one on this.

I wouldn't drive more than one axle on this. Any more and you lose efficiency and I'm sure you'll want it to move as easily as possible.

[ygmir]

While I think the skateboard style lean steering idea is brilliant, I think with four people on it you'd have a really hard time maneuvering.
You have to either have a differential or just simply drive one wheel and let the other freewheel - which is how most adult size tricycles work - or it won't want to turn.
I've lost the info, maybe Tiahaar has it, there's a perfect little diff available from an industrial supply place. But really, you don't benefit much if any from using one on this.

I wouldn't drive more than one axle on this. Any more and you lose efficiency and I'm sure you'll want it to move as easily as possible.

agree here. you could set up a double bearing ATV axle with the drive sprocket, and run the other side with forks (bogey), like a front end. same on the front. single drive, if you have good traction (atv tires) is plenty, and steering is way better/easier.

[Elliot]

...
...there's a perfect little diff available from an industrial supply place.
....

Northern Tool, online, 1" shafts, about $130 last time I bought one. These are commonly used in Kinetic Sculpture Racing. They require very precise mounting in four bearings, lest they break from constant flexing. The overall width is around three feet, so at least one shaft would need to be extended -- which is perfectly doable with a long split shaft-collar.

I would consider using a diff only for a one-person vehicle. With two or more people, separate drives have proven preferable -- and only partly because diffs like the one from Northern Tool are open diffs; no "posi-traction".
With separate drives, you can vary speed/power to aid steering -- or even do all the steering.

[Elliot]

.











This is not a suggestion – merely a brainstorm.

(And a mechanical engineer or draftsman… I’m not! LOL)

Diamond layout of four golf cart tires, 18 inches tall and 8 inches wide.

Platform 8 feet by 4 feet 6 inches.

The two tires in the middle of the wheelbase do all the work. The other two swivel freely.
Note the swiveling tires are a tad higher than the driving tires.

Separate drivetrains. Vehicle steers by varying human input side-to-side.
Thus no steering mechanism needed.

With no form of freewheeling in the drivetrain, inputs also serves as brakes.

So…. Two identical very-simple drivetrains, no other moving parts.

[Elliot]

Drive (gear) ratio changeable by re-positioning crank pin. Estimated time five minutes.

Similarly changeable position on pump arm. Estimated time 30 seconds.

Length of connecting rod... changeable to match crank and pump-arm -- two pieces telescoping, with row of holes and a quick release pin. Estimated time 30 seconds.

[BoyScoutGirl]

Ideas seem to be converging. Elliot, your idea is blissfully simple, and the illustrations really help this non-mechanic understand. Thank you!

We like the "tank style" steering - i.e., two sets of pumps and drive-trains, turn by pumping one set faster. More fun for four drivers, much crazier for two.

However, my co-conspirators are fairly set on the look of two axles, rather than three. Captain Goddammit mentioned that driving two axles from one pump input makes everything less efficient, but it remains an option. All four wheels would be driven, the left two by the left pump and the right two by the right pump. I guess that's technically four axles, as we couldn't have the left and right wheels on the same axle in this scenario. With the wheels fixed and turning accomplished by a left-right difference in pump speed I expect we'd have to fight a lot of drag. For clarity, let's call that plan C:


Variations on plan C:
Plan A for "probably Atrocious:"
Having the pump on the left power the left rear wheel and have the left front wheel free, while the right pump powers the right front wheel with the right rear wheel free would be a bad idea... wouldn't it? It seems like the fixed, powered wheels would cause a lot of drag:


Plan B:
So then I thought, maybe a combination of plan C and Elliot's idea?
We might be able to somewhat disguise the big tires in the middle by tucking them in the middle (this reduces mechanical advantage in turning, I believe) and drawing the eye to the unpowered front and rear BMX wheels. Two front wheels would be on a truck/bogie with limited range of swivel; same with the rear wheels:

[Elliot]

.


For... purely for appearance, you could use dummy wheels. These could be lightweight bicycle wheels. They would rest on the ground, but carry no weight -- or even be held up a tad to minimize scrubbing in turns.

The Lady Sophia, above, has such dummy wheels. The golf cart wheels are supposed to be covered by a skirt around the bottom of the golf cart.
The rear dummies rest on the ground. They are close enough to the real rear axle that they don't scrub much on turns.
The front wheels -- as you can see on the near wheel -- are partly held up by springs. They touch the ground enough to spin. They ought to have no tread at all, to minimize scrubbing resistance in turns, but it didn't matter with golf cart motor power.

As you correctly predict, your Plan C would not turn worth a hoot. With all four tires remaining in a square configuration, they would scrub something fiercely in order to turn. It would work fine for very gentle turns, but a typical BRC street corner would be too tight for it.

Plan A is more realistic, but the two drive wheels would still scrub quite a bit.

What allows Plan Elliot to turn, is.... Take a thumb tack or sewing pin or some such and stick it thru the center of the drawing, into the kitchen table. Now rotate the drawing around the pin. Plan Elliot will do that fairly well, with the main wheels rolling in opposite directions. Now stick the pin in the center of plans C and A. How will those drive wheels move across the kitchen table?

Plan B may work if you place the pivot points ahead of the axle, so you get a caster effect. But with the pivot points in the center of the axles, they will be free to do "whatever". One wheel encounters a small dust-dune, and the axle turns 45 or 90 degrees. Downside to Plan B with caster effect, is that the vehicle could not move in the other direction, as those axles would try to turn 180 degrees.
A better Plan B would be to use four caster-wheels -- essentially Plan Elliot with four casters instead of two, for the two-axle appearance. Hide the drive wheels with flat black skirts. This could also move in both directions.

[Elliot]

2006, my first Playa-specific vehicle. It worked tolerably well so long as the ground was hard. But it consumed bicycle wheels.

[Captain Goddammit]

None of those steering plans are going to work very well. Some them, not at all! Scrub-steering is going really to suck when you try to ride the thing.
If the co-conspirators are set on two axles, you're better off only driving one of them, I don't see where there's anything to be gained by driving both but extra drag and mechanical complexity. It's either going to have one fixed drive axle and one steer axle, or two steer axles and you're in for some fancy engineering to drive one of them. And more inefficiency through the gears you'll need to make it work.

[BoyScoutGirl]

Thank you for breaking it down to the simplest level for me, folks. You should see the rubber-band and index card mock-ups I've been playing with. It now sounds like it's time for plan E (for Elliot):

Plan E has the benefit of simplicity: the axles being driven are directly below the hand pumps. This plan also has the quirk of needing two pumps, left and right. Steering is accomplished by the difference in pumping speed between the left and right pumps. Fun for four riders, complex for two!

For this plan, would we want to limit the range of swivel on the casters, or allow for full 360° pivoting? Would we want to include a spring-type connection so that the left and right wheels are roughly in the same alignment? I know the one wonky wheel on the shopping cart always drives me crazy! This is, I suppose, why everyone recommended only one caster wheel in the first place...

Remembering the big stone carousel from Center Camp several years ago, I do worry about dust getting into our casters and making them stick, particularly with a lot of weight on them.

Further, this plan allows for the narrower wheels to make a useful appearance, provided we can obscure the drive wheels. An alternative would be to use cosmetic wheels.

An alternative: Plan F:
Captain, I believe this is what you're saying:


Plan F has the benefit of two axles, only one of which is driven and one of which is steered. If I'm reading the comments above from Tiahaar and Ygmir and others correctly, the suggestion is to power one wheel, and not the whole axle... I think? That's very counter-intuitive to me. In an actual handcar, the pump has a rod that turns a sprocket that turns the axle either directly or with a chain. Would be the advantage of only powering one wheel versus the whole axle be the lack of a need for a differential because the second wheel is free-spinning?

Any preference for front vs. rear-wheel drive? It seems to me that because the drive wheels are fixed and will drag/scrub, we'd want them in back.

[ygmir]

one drive wheel makes steering a lot easier, because you don't have two driven wheels, rotating at identical speed,, trying to push you straight, with the advantage of being spread and thus having leverage over the steering wheels.

I'd think you could have a solid axle, attached to one wheel via a hub and a drive sprocket, the other wheel free spinning, maybe by use of a bearing for it to ride on, or a front end hub sort of thing?

[BoyScoutGirl]

one drive wheel makes steering a lot easier, because you don't have two driven wheels, rotating at identical speed,, trying to push you straight, with the advantage of being spread and thus having leverage over the steering wheels.

Makes sense!

[Captain Goddammit]

In Plan F, the wheels will not drag or scrub in turns at all, unless you have a solid axle connecting the left and right wheels. You don't want that, for that reason.
Adding the expense and complexity of a differential will solve the scrubbing-in-turns problem, but so will just driving one of the wheels.
The differential only effectively drives one of the wheels at any given time anyway. Raise one wheel in the air (either one) and it will spin while the one on the ground will do nothing.
Actual rail pump cars don't have to make turns like your cart will. That's why they can get away with a solid drive axle.
If it's going to have a designated front and rear to it, make the front end steer and the rear the drive axle.

I'd do the three axle layout, but you've said your crew wants two axles.

Plan E will work, but it won't work well. It'll be hard to control, hard to go straight, and hard to turn.
Every time the left side and right side pumpers get out of sync, the thing will veer one way or the other. Have fun with that!
If you did the Plan E layout, I think you'd be infinitely better off with the steering axles actually connected to a steering system of some sort rather than relying on scrub-steering with the middle axle.

[Elliot]

The wonky wheel on the shopping cart brings us to the matter of suspension. The shopping cart has no suspension. Theoretically all four wheels are mounted in the same plane, and on a flat floor they should carry the same weight. In reality, a tiny twist in the cart, maybe from once being loaded with 14 six-gallon cases of water and three screaming kids jumping on top, will leave the cart largely sitting on three wheels. The fourth wheel barely touches the floor, maybe only part of the time, and the wonky-dance begins.
Or a bearing wears out -- any one of the six bearings (four wheel bearings and two swivel bearings) -- same result.

Take a look at a RhoadesCar -- mass produced in Tennessee. Plenty of them on Playa. There is a tiny rubber "spring" on each front spindle, but it amounts to nothing. So a RhoadesCar travels on three wheels most of the time. The weight of the riders is in the rear, so it's always a front tire that's in the air, which does no harm (but you can feel it in the steering wheel).

The trouble begins when it's a drive tire that comes off the ground -- and the two drive tires are connected by an open differential. The wheel in the air will spin, and the vehicle goes nowhere. Same if one drive wheel is in a dust-dune -- or on snow in Minnesota in January. One wheel spins, vehicle stands still. This is why some cars have a limited-slip differential -- known by many names and also of many designs. Posi-Traction comes to mind. (Never buy a car without it!)

The open diff problem evaporates when you power each wheel independently. In fact, you gain the advantage of being able to aid steering by adjusting the power to each wheel. More power to the outside wheel will help the vehicle turn. (This can be a great help in Kinetic Sculpture Racing, when, for example, making a sharp turn on sand.)

The RhoadesCar has no diff, but it does the opposite of helping the steering. The axle is solid from side to side, but each wheel is mounted on a freewheel, which allows it to spin faster than the riders are powering it. So in a right-hand turn, the left wheel coasts faster, which prevents scrubbing. (Remember that the outside wheel in a turn must travel farther, so it must spin faster.) That may seem fine at first glance. Buuuuut.... Now it's the inside wheel that does all the forward propelling, and... yes, it pushes against the intended direction of turn. RhoadesCars vigorously resist sharp turns. I fixed mine in 20 seconds with a Sawsall -- by cutting the axle in two. Now it has two independent drives.

Adult tricycles, such as are popular in BRC, drive only one rear wheel. Same issue. They turn very well to one side and resist turning to the other side.

Where were we? LOL

No need to limit or connect the swiveling of the four caster wheels in Plan E. They will each take the path of least resistance.
Plan E can also rotate in place, as mentioned when we stuck pins in the kitchen table. The casters will all turn 90 degrees and around we go.
And you can reverse travel-direction. There will be a struggle as the casters fight their way around 180 degrees, but after that all is normal. You may have to label the ends of the platform to know which is which -- though I don't know why you would need to.

The most all-conquering vehicle I have built has four wheels and four riders. Each rider propels his own wheel. Four completely independent drivetrains.
And I didn't even mess with any sort of steering on either axle. I simply built the thing in two halves, and connected them with a pivot in the middle -- like some earth movers and such big equipment. The center pivot can also twist -- that's its "suspension", keeping all four wheels always on the ground. The silly thing is bloody near unstoppable.



Late addition:
Ygmir and Goddammit are right about vehicles insisting on going straight if two wheels on one axle are locked together. You can probably get away with it to some extent on Playa, but on pavement it will absolutely refuse to turn. Beginning Kinetic builders discover this the hard way, almost every time.

There are vehicles that consist of two wheels side by side -- and nothing else. Picture of one of them below. It has two riders, and each drives one wheel. Beginners usually zigzag quite a bit, but after a few minutes they master it remarkably well.


Unwieldy 2 by Dave Hershberger. Photographer unknown.

[Elliot]

Oops. Good night.

[Elliot]

...
I'd think you could have a solid axle, attached to one wheel via a hub and a drive sprocket, the other wheel free spinning, maybe by use of a bearing for it to ride on...

That's exactly how the common adult trikes are built.

[Elliot]

The aforementioned Contraption by Le Chat Noir and Karine.

[Elliot]

.
For late night amusement....



Doing it wrong.