Steampunk vehicles

[Alexander Watt Babbage]

I'm working on a Humber 3-speed for Tweed rides, but these take things
to another dimension: http://www.ridelust.com/adventures-in-anachronism-10-steampunk-rides/

[Don Ricardo Penrose-Marconi]

I'm building a vehicle for Burning Man and if I get so far as to give
it some real aesthetics, then I will echo this theme. Perhaps after
Labor Day I will be ready to look for style pointers. At present I am
locked in mechatronic purgatory with the thing, i.e. I just want it to
run. (Won't be at any SSS meetings until then for this reason.)

(I like how the Society initials sound like a pressure valve actuating
steam release.)

On Jul 28, 5:40 pm, Alexander Watt Babbage <dh...@sbcglobal.net>
wrote:

[Alexander Watt Babbage]

What kind of mechanical difficulties are you encountering? Or is it
electronic motor controls?

[Don Ricardo Penrose-Marconi]

At present, learning that bicycle front ends do not make good
casters. 'Pears I need to increase the trailing distance to get some
stability. And needless to say, the solution space is constrained by
the state of construction. Conceptually, the thing's a motorized
wheelchair and the front non-driving wheels shimmy like the dickens.
Once the shimmy is solved there will be other discoveries, I'm sure.

[Dack Patrick O'Connor, Esq.]

A slight negative camber of just a couple degrees may solve your
shimmy.

On Jul 29, 4:37 pm, Don Ricardo Penrose-Marconi <drtee...@gmail.com>
wrote:

[Don Ricardo Penrose-Marconi]

By negative camber do you mean, in front view, the wheels will lean
out somewhat, or lean in?

On Jul 29, 4:40 pm, "Dack Patrick O'Connor, Esq." <b...@homeacres.net>
wrote:

[Don Ricardo Penrose-Marconi]

No, I see, you mean so the wheels lean in a bit. Can try that (or try
to try that, whole lotta welding going on).

On Jul 29, 5:21 pm, Don Ricardo Penrose-Marconi <drtee...@gmail.com>

[Alexander Watt Babbage]

What you need is the appropriate caster/trail.(tire contact distance
behind the steering axis). The diameter of the wheel makes a big
difference.

For small diameter, non-driving wheels, the usual method is a vertical
pivot axis with a considerable offset to the back for the wheel axle.
Look at the carts at home depot for measurements for the amount of
offset. The correct trail depends on the size wheels you are using.
Increasing the weight on the caster should reduce flutter. If yours
is a four wheel vehicle - introducing some suspension either at the
front or back will help keep weight evenly distributed to prevent one
wheel flutter (which will cause the vehicle to turn unpredictably when
it occurs) The tendency and frequency of the flutter is speed
dependent - I assume your vehicle doesn't go very fast = a shorter
trail may shift the frequency of flutter out of your speed range.

For mild flutter packing the axis bearings with a very stiff grease
might provide enough dampening forse.

[Don Ricardo Penrose-Marconi]

Is there a rule of thumb for trail vs diameter? T >= D/2 or something
like that?

I am going to use 20" BMX wheels because the frame won't allow me to
set 24" wheels back very far. Even so I will probably only get a two
or three inch trailing distance. (Thus hoping my formula above is
wrong.)

Weight will be significantly on the rear (drive) wheels. Will see how
I can shift the load forward, but am severely limited by the swing
room needed by the front wheels (not putting the carriage base above
them).

Max speed should be in the 5mph range, maybe a little higher.


On Jul 30, 8:52 am, Alexander Watt Babbage <dh...@sbcglobal.net>
wrote:

[Alexander Watt Babbage]

Bicycle front end geometry is optimiized for steering and self-
steering - the opposite of what you want. Bicycle steering axis
inclination is in the range of 70-75 degrees. You want 90 degrees
(vertical) with all of the trail behind the vertical spindle.

I'm pretty sure that trail is usually in the range of 1 to 6 inches
from very small to about 30 inch wheels for practically any
application. The key is to achieve stability and self alignment in
your speed range. A shimmy is the mass of the wheel assembly itself
oscillating around the spindle axis. Friction in the bearings and the
tire contact will help dampen the oscillations up to the point that
speed introduces too much input forces for them to control. Trail
provides a lever arm for the forward momentum of the vehicle to act
against the road surface. You have to have some trail in order to get
a self-centering effect. Too much trail and speed makes for an
excessively strong self centering effect that causes the mass of the
assembly to overshoot center and create a self-sustaining shimmy at
certain speeds.. At the walking speed you are talking about very
small changes in the trail should make big differences in the
oscillation dynamics. Tuning of a particular system has to take into
consideration (at minimum) the polar moment of inertia around the
spindle axis, trail, design speed, and dampening characteristics of
the tire (somewhat variable with inflation.) More weight on the wheel
and less tire pressure will give you more dampening friction.

Moving the rear wheels further back from the center of gravity of the
entire vehicle will transfer more weight to the front wheels. I don't
know exactly how much trail you need with a 20 inch trailing wheel,
but your speeds are low enough that it should be possible to try some
experimenting. You want to avoid making the front wheels any further
forward of the CG than necessary. Changes of as little as one
centimeter can make a noticeable difference. I will guess it should
be in the 2-3 inch range.

Take a 20 inch bicycle and remove the bars. Then turn the fork around
so you have the offset to the back, hold up the rest of the bicycle
and roll it along on the front wheel. Raise the back of the bike
until the headset is near vertical. Experiment with various angles to
see what works. Although your actual spindle axis should be vertical,
you can change the effective trail by this method. Once you have a
rough idea of what works - you could build something more precise to
experiment with - still easier than getting it wrong on your vehicle.

Here is a very technical discussion (beyond me) for what you can glean
from it: http://www.rehab.research.va.gov/jour/00/37/3/kauzlarlich.htm

Some on-line discussions of tail-wheel airplanes suggest that the
pivot axis needs to be very SLIGHTLY inclined forward (same direction
of angle as a bicycle.). I suspect that solution is unique to the
airplane suspension and would not be an advantage for you.

[Don Ricardo Penrose-Marconi]

This is great input. Thank you very much. I saw the Kauzlarlich
article upon searching earlier this week but couldn't take the time to
translate it into a practical guide for my app. Not sure I would have
been able to anyway. I will modify the metal to experiment with
trailing distances. I also have to make mods so the turning spindle
is normal to the ground, because initially my assistant had a feeling
a slightly negative caster angle would help. 'Pon reflection, not so
sure. Too late to leave as is because straightening that on one side
was our first post-flutter experiment.

We're engineers more in the Maxwell and Faraday realm than that of
Newton.

It's become much more of a science project than intended but I'm
confident that when we are successful and mount one of those
compressed-aether cannon, we'll give the Zulus something to sing
about.

Thanks again!

On Jul 30, 6:05 pm, Alexander Watt Babbage <dh...@sbcglobal.net>
wrote: