[python] Re: Python Stability
- From: UH-13P <uh13p@xxxxxxxxx>
- To: "python@xxxxxxxxxxxxx" <python@xxxxxxxxxxxxx>
- Date: Mon, 2 Jul 2012 09:37:49 -0700 (PDT)
I would add some considerations from reading the Python list. The reports of
"easy to learn" seem to me to have come from the people using 20" front or both
wheels. The small wheels indirectly result in less negative trail and designs
with the rider weight relatively higher in relation to the bike. When
comparing the dimensions to low racer and diamond frame experience both would
lead to a more stable and an easier to ride Python.
Secondly, dynamic stability is a result of the system and not a single part.
PIO was mentioned in an earlier post and results when feedback is out of phase
with the required correction. Leg steering a large pivot arm vs arm steering a
pivot nearly inline with the wheel will have greatly differing time constant in
the feedback loop. Any given rider and bike combination will be different. An
example might be 5 of us in single file at the same speed with mountain
bikes when one of the 5 cocks the front wheel and goes over the handlebars.
The other 4 don't notice anything other than a slight turn. The 1 rider bike
combination resulted in a steering correction not matching the condition
faced.
________________________________
From: George Durbridge <gdurbrid@xxxxxxxxxxxxxx>
To: python@xxxxxxxxxxxxx
Sent: Monday, July 2, 2012 6:45 AM
Subject: [python] Re: Python Stability
On Sun, 2012-07-01 at 22:25 +1200, Peter Clouston wrote:
> George, if you read my post carefully you will see that I am not
> claiming that wheelflop is the reason for a DF bike's stability, just
> that the small amount of positive wheel flop of a standard DF is said to
> have a small beneficial effect on stability.
Don't take any of this personally. Discussions here are fairly robust,
and your points are genuinely interesting. I did read your post
carefully, and I know that is what you said. You just didn't follow the
argument through.
To enlarge: trail, with or without moderate rake, contributes to
stability, at certain speeds. The Papadopoulos and Ruina equations and
the bike-rolling-down-a-corridor tests both show that there are windows
of stability, speed ranges in which a given bike is stable, and outside
which it is not. (I am assuming here that what contributes to stability
of a riderless bike assists a rider to control the same bike.)
I'm no good at differential equations, but it's fairly clear that a bike
with wheelflop is unstable at zero speed and very low speeds, and the
more wheelflop the less stable, but at higher speeds may become stable
because corrective forces which increase with speed come into play. A
python is more unstable at rest than the usual bike (if you prop it
against a post, it falls over) because (a) it has more wheelflop than
the usual bike, whether upright or recumbent, (b) the mass which rotates
around the steering axis is greater, and (c) the centre of gravity of
that mass is situated much further from that axis, than with the usual
bike. That suggests that a python may need a higher speed to become
stable than the usual bike, which is consistent with my experience.
All of that suggests that pythons may be hard to learn to ride, but that
the difficulty is mainly in starting from rest, and the bike is rideable
once under way (which is what I see in teaching the neighbourhood kids
to ride upright bikes). My own experience hardly got that far, but
that's what others report. It probably follows that you don't have to be
extraordinarily adept to ride a python, as Peter is inclined to think,
but that most people could learn to ride one with either (a) experience
on low-racers or (b) two pairs of leather gloves and a month's practice,
which I seem to recall was Jurgen's summary. (Take-home question: is
there a third way? Python trike?)
Patrick's point that every python is different is well taken, and may be
one of the reasons why Dirk doesn't see the same deterioration of
stability at higher speeds that others do. (Other reasons are that Dirk
has built more pythons than most, and that he understands the dynamics
of pythons better than most.) But so many riders report instability at
speed that I think some pythons must be susceptible to it. Vi suggests
it's due to PSI, which may be supported by Esko Meriluoto's observation
that the Hipparion was more stable with a steering damper fitted. I
think it's probably positive feedback due to negative trail; a bit like
the unstable steering on a billy-cart (antipodean reference for Peter)
or on a Christiania cargo trike. The two explanations are consistent
with one another, as PSI and road irregularities are the two obvious
factors initiating positive feedback.
> In fact such a bike would
> still be stable with zero wheelflop, provided that the trail remained
> about the same. (Zero wheelflop requires a vertical steering axis,
> which on a DF makes providing adequate trail, turning clearance between
> the front wheel and the toes and having the handlebar in easy reach all
> at once very difficult.)
Agreed - rake seems to be an ergonomic requirement, not a dynamical one.
Tony Foale adapted a BMW motorbike to have no rake but adequate trail,
and said it was fine. The original Windcheetah had negligible rake, but
what works for trikes need not work for bikes.
============================================================
This is the Python Mailinglist
//www.freelists.org/list/python
Listmaster: Jurgen Mages jmages@xxxxxx
To unsubscribe send an empty mail to
python-request@xxxxxxxxxxxxx
with 'unsubscribe' in the subject field.
============================================================
Other related posts:
