I'm finally getting round to writing up my current thoughts on a TTW HPV
so I can invite comments, suggestions, corrections, etc..

My interest started with a discussion about recumbent bikes, and thinking that,
while it would be fun to have one, for riding in town with frequent stopping
and starting, a conventional upright bike was easier to put a foot down on.
(At the time I lived and worked in Cambridge (UK). I still work there, but now
live fifteen miles away.)

Obvious answer - a trike. But then we lose some of the advantage of being able
to filter though traffic, or fit though barriers on some cycle routes.

My first thought, of compensating by fitting a modest electric motor to bring
the trike up to 30mph to keep up with motor traffic without having to get too
fit, was ruled out by finding the Electrically Assisted Pedal Cycle rules here
say a motor can't assist above 15mph (unless the vehicle is registered, taxed,
insured, etc.). (And it looks like the rules will change soon to need a Single
Vehicle Type Approval examination, which could be expensive, but I guess
something similar would be needed for insurance anyway (SVA used not to apply
below 410kg unladen).)
http://www.roads.dft.gov.uk/conindex.htm

So we want a narrow trike. There are trikes which are not much wider than a
bike with panniers, or even no wider, but to be stable they need to be low,
which also has downsides in city traffic (though a friend of mine used a
Windcheetah for much of the same route I would now use (he did it in the
opposite direction though, against the heavier flow)).

So, how do we get a narrow stable not-low trike? Make it lean.
Mitch must have been reading my mind (or more likely looking at who browsed his
web site), since he sent me an invitation to join the tilting group, so I could
learn from lots of people with similar ideas.

The main requirement is that it's got to be cheap and relatively easy for me to
build (at least for an initial prototype).


On to design:

Wheel configuration - delta or tadpole
--------------------------------------

I feel the "right" configuration is 2 front. Compare a Grinnall Scorpion or a
Morgan with a Reliant Robin or a Bond Bug, or a Windcheetah or a Trice with a
Kettwiesel. (Though I do like the look of the Carver.)
http://www.3wheelers.com/grinnal.html http://www.3wheelers.com/morgan.html
http://www.3wheelers.com/robin65.html http://www.3wheelers.com/bond.html
http://www.windcheetah.co.uk/, http://www.ice.hpv.co.uk/,
http://www.kinetics.org.uk/html/kettwiesel.html

Understeer at the limit when we hit the lean stops sounds probably good, and
having two braking wheels when weight transfers forward would be good too
(and an HPV can brake much harder than it can accelerate). And all the fully
faired trikes I've ever seen photos of have been tadpole, not delta - it fits
better in a fairing/body with a nicely tapered rear.

_But_ single front wheel steering will be much easier to build (without Harry's
"tubing jungle"), and this isn't going to be a racing machine.

The Tripendo (http://www.tripendo.com/) with its two front wheels and all
wheels leaning looks lovely, but it is heavy (for an HPV), complicated, and
expensive. (A more conventional cross frame trike with the front wheels
non-tilting and the main joint replaced by bearings to allow the front-rear
main member to pivot is another possibility, but still a lot harder than just
using a standard bike headset and fork.)

So I'm fairly convinced that one front is the way I'll actually go.


Wheel configuration - how many leaning
--------------------------------------

I really want to lean all three wheels (especially since they will be bike
wheels
and tyres). But conventional trikes make bike wheels handle side loads, even
uprights with full-size wheels, so it isn't actually necessary.

There seem to be three obvious possibilities:
1) rear parallelogram, tilting hubs on "uprights" at end (like the front
of a Tripendo, without the steering bits, or the back end of
the trike in http://sunsite.anu.edu.au/community/ozhpv/huff/huff15.pdf
and http://sunsite.anu.edu.au/community/ozhpv/huff/huff16.pdf

2) parallel swingarms, like a Calleja, using rear triangles from a couple of
suspension bikes (but with springs and dampers acting on the linkage
between the two, so they tilt freely - or possibly discarded, since
the units on a cheap child's suspension bike might not be worth using).

3) Non-tilting rear end, like a Flevotrike.

With rear wheel drive and driveshafts (which I'm told are axles in US English,
with a US driveshaft being a UK propshaft), it's easier if the rear doesn't
tilt (no wide-angle CV joints needed), but then we have to deal with chain
twist between the tilting pedals/chainset and the non-tilting rear end. Since
part of the point is to have the seat higher than a typical trike, we can have
a fairly straight chainline without intermediate rollers or jackshafts to get
under the seat, so that needn't be a problem.

If I use front wheel drive, like a Zox (http://www.zoxbikes.com/) or Flevobike
(http://www.ligfietsshop.nl/nl/tempelman/bike_nl.htm), then a rear parallelogram
is probably as easy to make as a non-tilting rear, and more compact than the
swingarms (while I want a higher riding position than a typical trike, I think
the rear wheels will still be more behind than under the seat).

If I go for the parallel triangles, then without either a differential or an
opposite handed freehub for one side (which exist, but are unusual and
expensive), I think I'm just running a chain on one side, but that's not
unusual on 2R trikes. (Chain running down the offside of the nearside triangle
(down the right of the left triangle, viewed from the rear), from the nearside
of the main frame (with tandem-like crossover so front chainset is still on
normal side so pedals don't unscrew).)
Unless anyone can suggest a source of small light cheap differentials.

Something like http://www.bikesmithdesign.com/temp/Jackshaft.jpg with
freewheels on an intermediate shaft would work for the non-tilting rear,
but be trickier with either of the tilting schemes. If the jackshaft is
further back than the rear axles, it even reduces the twist in the chain.

Of course a single tilting rear wheel would be the simplest, but having decided
on a single tilting front wheel, that would make it a bike, and then it
wouldn't stay upright when stopped without putting a foot down. Unless I make
it a bike with retractable stabilisers instead, like a Peraves Ecomobile
(http://www.peraves.ch/ndexe.htm), but then it gets off topic here, so we'll
ignore that for now. It is tempting though. (And I love "embarrass the latest
supermotorcycles in the swervery with it").


Front or rear wheel drive
=========================

Front wheel drive doesn't allow large steering angles if it relies on chain
twist, but it could use a universal joint - see http://www.hpv.on.ca/ghuvj.htm
or http://www.stitesdesign.com/hpv_chameleon_fwd.htm

(Front wheel drive on a two front wheel trike is possible but rare, and tilting
will make it trickier - http://www.aha.ru/~ykpro/index.htm/)

Front wheel drive with a non-tilting rear end with hinge at the front would
allow easy use of a load space between the rear wheels, but that's not needed
for commuting, and longer/more hinges would be stronger (something like the
Carver where the tilting part of the body reaches behind the rear wheels
instead of just being stuck on the front, though the Carver has much higher
forces to deal with than an HPV of course). A single hinge (using a headtube)
would be easier to align, and to remove for maintaining, though.

I'm not going to be climbing any steep hills, so the front-wheel-drive's
tendency to lose traction in that case (with weight more on the back) isn't a
problem. On the other hand it would be nice if the design was usable for
touring too.

But so long as chain twist isn't too bad rear wheel drive means using more
standard bike parts, (for example, it means using a standard suspension front
fork is possible). (Mounting all the gears on the boom with a single-speed
front wheel might allow front wheel drive with a standard fork, but even
single-speed rear hubs are normally wider than front ones, though only
slightly. The boom starts getting very cluttered then though).


Lean type
=========

Total free lean is out, since we started off with not wanting to fall over when
stopped without having to put a foot down. Active/automatic lean is out (power
consuming and complex). So it's free lean with a low speed lean brake, or it's
forced lean.

It would be boring just to copy a Flevotrike's fixed ratio lean-steer, so
that's ruled out. If I get to try one and like it, I'll change my mind, but it
just looks wrong to me. Lean and steer should be independent so the ratio can
vary with speed (or have a ratio that changes with speed, but that's tricky on
an HPV).

I want to be able to work the forced/locked lean at low speed while giving
hand signals, so a seperate lean lever like the Tripendo is out, unless
I've misunderstood it (or there is a lock as well).

Having a full-time lean control that can override free lean but normally has
no significant force applied to it seems easier than a seperate lean lock,
but an extra thumbshifter linked to a lean lock is possible. I've ruled out
lock when brakes are full on or a coaster brake operated lean lock, so it
can be kept on for low speed tight turning (the sort of thing that on an
upright bike you would just pick up the bike under you), and doesn't lock
if you brake in a turn.

If I don't have a seperate lock, underseat steering that controls steering by
yaw angle and lean by roll angle seems likely. Above seat steering gives a more
aerodynamic position, generally, and we could use tiller to steer and twist to
lean, but I think the mechanism wouldn't be as straightforward.


Lean axis
=========

Low is good. Actually having the instantaneous roll centre at ground level with
a four-bar linkage is possible, but does mean it shifts from side to side, and
means more bearings to align. Just keeping it low is probably adequate. With
the rear wheels relatively close together, and not being designed for off-road
use, and having transmission components fairly low anyway, we don't need to
worry a lot about ground clearance in the middle.

I'm not going to be doing fast downhills on this leaning over at a 45 degree
angle, but ideally the machine should allow for such a lean. See
http://www-math.science.unitn.it/Bike/Countries/Europe/Tour_Reports/Tour_of_the_\
Alps_Gallery/tiretest.jpg

See the diagrams on http://www.wannee.nl/hpv/variatie/e-var-6.htm
Now look at http://www.stitesdesign.com/hpv_chameleon.htm, and I think
the lean axis there is too high - there's really not much lean before
the line of the front wheel comes outside the back wheels.

The Chameleon is a bit similar in overall layout to what I'm currently thinking
of (though it is front wheel drive), but I want to be a lower and more laid
back, just not as low and laid back as a fast trike like the Windcheetah.


Wheelbase
=========

Need to try a few bikes and trikes to compare. Almost all new European
HPV's seem to be short wheelbase (or convertible between long and short,
like the Burleys (http://www.kinetics.org.uk/html/burley.html), but
long wheelbase does seem more common for delta trikes.


Wheel size
==========

20" all round I think. I don't want more than one size, and smaller wheels are
stronger in side load and easier to fit everything in. Downside is I'll need
intermediate gearing in somewhere to get the speed back up, but many trikes
have that even with a larger rear drive wheel.


Suspension
==========

Possibly just rubber mountings for the seat initially. Even some very
cheap 20" wheel children's bikes have suspension these days, but it
isn't very good, and I'm much heavier than a small child. With a
single front wheel in a normal fork adding front suspension will
be a straight swap, the back still needs more thought.
http://www.hpvelotechnik.com/produkte/federgabeln/index_e.html

--
Alan Braggins mailto:armb@... http://www.ncipher.com/
nCipher Corporation Ltd. +44 1223 723600 Fax: +44 1223 723601