[python] Re: Lean Steer Python/Hipparion Trike

  • From: dirk.bonne@xxxxxxx
  • To: python@xxxxxxxxxxxxx
  • Date: Tue, 09 Jan 2007 18:48:51 +0100

Henry Thomas wrote:
> George Durbridge wrote:
>>> It is very exacting and hard to get your head around at times.   
>> Agreed - I am still trying to think out a lot of it.
> Actually with the new simulation geometry it is making a lot more
> sense, the funny thing is that only one variable really matters, all
> the other ones have a marginal impact. I will detail this when I am
> finished drawing up some diagrams to explain it clearly.
>> Thanks for those references.  The previous trikes all used a rocker arm
>> moving in a vertical plane about a horizontal pivot, but your current
>> design has the rocker arm moving in a horizontal plane about a vertical
>> pivot: did you make that change to achieve self-centring, or for
>> construction reasons? 
> Much like the python pivot, it doesn't mater how height it is off the
> ground, as long as it pivots on the axis at the correct angle and
> trail. Likewise with this you can rotate the mechanism around the
> pivot axis and place it where it is easiest to construct. It did I
> guess effect the self centering effect in that by separating the
> leaver from the arm I was able to simulate it better.
>> Either way, had you thought of substituting a
>> stiff leaf spring for the rigid rocker arm, to allow some suspension
>> movement, other than pure roll?
> This is a nice idea, but might be a bit heavy. I am looking at placing
> an elastomer on each end of the rocker arm.
>> By leaning the wheels, we are loading them with cornering forces they
>> are designed to resist, unlike trikes cornering with upright wheels.
>> And, unlike so many leaning trikes, you do it with all three wheels.
>> While that has to be a good thing, it really affects how long the wheels
>> will last, rather than the cornering forces we can generate in the
>> meantime.
> I disagree, the thing I didn't do before but have done since, is draw
> a line collinear with the riders CoG to the ground. It turns out that
> the line is neutral, just like a bike. It doesn't move out or in as
> the trike tilts. I was surprised to discover this as I thought it
> looked like it moved marginally out, but it was just an optical
> illusion. So if the line doesn't move from center, and the rear wheels
> are parallel with it, then they behave just like a bike and only have
> a radial load, not a angular load, so I would not expect them to wear
> any differently. Even so, I have 48 spoke BMX wheels for the back with
> double wall rims and 14mm axles just in case ;-)
>> However, if the CoG neither falls nor moves inward when the trike leans,
>> have we improved the trike's cornering power at all?  The cornering
>> force it can generate (if the tyres and everything else are up to it) is
>> limited by the location of the CoG: the maximum cornering force in g is
>> equal to the tangent of the angle to the horizontal of the line from the
>> outer contact patch to the CoG.  Unless we can reduce that angle, can we
>> improve the cornering?
> I think this is where my earlier description might have been
> confusing. If you imagine a rider on a bike with the contact patch
> below them, as they lean they move in an arc about the contact patch.
> My tilting mechanism does just this, except at the same time the rider
> leans, they are also moving away collinearly from the contact patch,
> so that their actual height parallel to the ground is lifted ever so
> slightly. The effect is that the static force of this lifting will
> tend to self center the trike. However if the rider leans and moves
> their CoG to one side or the other, this shift in weight will over
> come the self centering effect. That is it should be easy to find an
> equilibrium yet easy to lean and tilt.
If I understand correctly the contact point of that axis stays in the
middle of both rear wheels. That is cool.

But why would you want the CoG to rise? What advantage would you have in
self centering for leaning. Except the self centering the behaviour is
more or less like that of a two wheeled bike, which has fine leaning
characteristics. I bet that the self centering of leaning will only
confuse the driver - or is it that small that it can be ignored?
>> Of course, all of this assumes you are strapped to the seat, not leaning
>> further into the corner than the seat: maybe the real point is that the
>> leaning seat will facilitate your moving the CoG into the corner,
>> without hindering you straightening up again.
> The other thing to know about my design is that the tilt angle is
> naturally self limiting. That is, it will only tilt 25-30 degrees and
> lock -- this is an effect of the geometry. The rider CoG never exceeds
> the inside wheel, so they will never tip over.
I think it is still possible. It can happen when the front wheel whipes
out. Then the CoG point can move outside of the triangle defined by the
two rear wheels and front wheel.

On my flevo trike (which also leaned but no leaning wheels) I sometimes
capsized on ice because of slipping away of the front wheel. AFAIR I was
not leaning at all. The whole thing just went away from under me.

But on the other hand, I guess you don't have that much snow in
Australia :-D



This is the Python Mailinglist


Listmaster: Jürgen Mages jmages@xxxxxx

To unsubscribe send an empty mail to 
with 'unsubscribe' in the subject field.


Other related posts: