[python] Re: some thoughts
- From: "25hz" <25hz@xxxxxxxxxx>
- To: <python@xxxxxxxxxxxxx>
- Date: Sun, 2 Oct 2005 22:44:24 -0400
First, the more I ride it, the more I think the python is an amazing design.
The feeling of slowly conquering the beast to make it rideable is extremely
satisfying for me - always gives me a big, wide grin when I think about what
it was like to start, and where I am now.
I would tend to agree with most of Jurgen's comments, and have a couple of
my own to add. I'm the kind of guy that has to "understand" how something
works. It's not good enough to know the answer, I need to know "why" it's
the answer too. Because of that . . . "trait", I tend to do a lot of
experimenting to find out why things work the way they do. So, onward . . .
- high speed coasting, I have found, is enhanced by continued pedalling to
keep light "weight" on the pedals while riding the rear brake. While this
is hard on the pads, the top speed slowly increases and I have hit 55 kph 2
or 3 times now.
- increasing weight on the FRONT also enhances stability. Right now, if the
bike leans to the left, the weight takes the pivot, and lets it fall to the
left, which moves the BB to the right, and increases the tendency to fall to
the left. The bike is inherently un-stable. If you add more weight to the
front, by the BB, now, if the bike leans to the left, the BB is heavier, so
IT falls to the left, moves the pivot to the right and the bike tries to
self-correct itself. This would be further enhanced by adding extra weight
BEHIND the rear axle. It would compliment the effect of the extra weight on
the front and add to the bikes desire to self correct. Now, the obvious
question is, what is the point of adding extra weight when you could simply
build a more conventional design recumbent, and eliminate the extra
counterweights? There isn't a point, aside from experimentation. The
python is rideable as-is, so the counterweights would simply be an
experiment, and the amount of weights would be determined by their distance
from the tires' contact patches in relation to the rider's weight.
- why does the python need to have two wheels the same size? Aside from
aesthetics, is there another purpose? I changed my rear wheel to 20" from
the original 26", and it lightened the bent by having a shorter frame and by
having a physically smaller wheel, plus it shortened the bike to make it
easier to transport and store. The next omni-python I'm about 60% through
building will have a 26"/16" set.
- for pivot angle evaluation, the current python I'm building has rod ends
(as I posted prior). It will also have an adjustable wheel-base. Right now
there are two (soon to be 4) pythons and 5 people who can ride them to
varying degrees. I am going to take each rider and have them ride the
python while I change the pivot angle from about 58 to 73 degrees,
inclusive, and starting from about 65 degrees, and have them simple rate
whether the angle change feels "better", the same or worse, in a general
term, than the previous angle. I'll also record traill, seat height and
wheelbase. Once the rider has hit their "sweet spot" for angle, I'll start
adjusting the wheel-base from about 38" up to 50", and again, record their
impressions. If they want to add more commentary about what specifically
they thing has changed, that's fine too. While I think the "sweet spot"
might end up being sort of rider specific, we might also find that all the
riders end up in a general area where the angles and wheelbases are close.
These impressions will be on the flat, and for angle changes on high speed
descents, I'll rely on Marcel and I to assess those :) I also put
suspension on the new python which can be locked out, so I can see what
effect that has on rider impressions as well. A lot of data to tabulate,
but once we get a good process, it should go along very fast. The angle can
be changed quickly with 2 wrenches, and the wheelbase adjustments are done
with two quick releases.
- the higher the seat, the easier it is to maintain your balance. The
practical limiting factor for this, I would think, is the height at which
you can not touch the ground with your hands anymore.
- the more upright the seating position, the easier it is to balance. Our
normal upright orientation has developed a strong sense of balance as well
as all the associated muscles. In a low, laid back seatnig position, the
body has to develope new and different processes for balancing and instead
of simply being able to tilt your head, or lean sideways, or shift your
hips, on a reclined seat, the entire upper body has to roll. While this can
be learned, it's still not as subtle a movement as in a more upright
position. When switching between my python and other more upright bents or
diamond frames, I am very aware of the differences between the different
bikes and notice that I am constantly "correcting" on the python while the
other bikes require far less rider input. Again, not a big deal - it's the
breed of the beast :)
- I think that BB to seat height has some kind of effect, but I am not sure
what, or how much. I could have (and should have) made the BB height
adjustable as well, but maybe for another python or maybe I can mod the new
one later.
- someone said that towing a trailer tends to make the python more stable
too. I would think that it would depend on how and where the trailer is
mounted to the python would determine whether it was a help or hindrance.
Something to definitely experiment with, but I'll have to build a new
one-wheeled trailer as I sold my little BoB clone.
- because hte python runs so well on the flat, and seems to take hills so
well (sometimes up to 3x as fast as I could do on my 42lb trike and the
python is only 5 lbs lighter) I think that I will use a 24" drive wheel,
with a 48/38/28 chainring and an 11/34 eight speed cassette. That will give
me about 20 to 105 gear inches, which is fine for hills and likely more top
end than I will ever be good enough to utilize. :) As it is, on some of
the hills I regularly ride, I had to use the granny gear with the 34 tooth
rear cog. On the python, on the same hills, I haven't had to use the granny
gear yet, and again, I'm only 5 pounds lighter than on my trike.
- both Marcel and I have had punctures on our pythons, and both were rears.
I am going to look into airless tires because at the speed you can ride on a
python, it's not a good thing if you can't hear the tire going down due to
the wind howling in your ears, until the handling goes haywire from the flat
and you wipe out.
For me, I want to see how close I can get to 60 degrees and how much it
helps on downhill performance. I'm also interested to see how pronounced
the wheel flop is as the angle gets lower.
I'll add the new python pictures to the page, and we managed to get an MPEG
of us riding the pythons. That will also be uploaded as soon as possible.
Sorry for the epic :)
> Hi all pythonauts and wannabes,
>
> weather is bad in these days. Good opportunity to contemplate
> about python steering geometry ;-)
>
> Now that we have around 10 experienced python riders here
> in this list, maybe we can enlarge the knowledge base about
> this unusual project.
> Looking forward to your opinions!
>
>
> First some of my observations:
>
> - the python feels stable as long as the front wheel is driven
> - coasting is easy upto 20 km/h for beginners
> - coasting is ok upto 40 km/h for more experienced riders
> - coasting upto 55 km/h is possible for skilled riders
> - more weight on the rear part enhances the coasting abilites
>
>
> And now some conclusions:
>
> - python steering geometry is different from normal bikes
> - main differences are the negative trail and self-centering effect
> - negative trail (NT) is a destabilizing effect while coasting
> - this effect increases (linearly?) with the coasting speed
> - NT is neutralized upto a particular speed by the self-centering effect
> - NT destabilizing effect can be damped by the riders skill
>
>
> So how to improve the performance?
>
> - improve your riding experience ;-)
> - practice high speed downhill coasting
> - try to move weight from the front part to the rear
> - evaluate the perfect steering pivot angle
>
>
> The last point of course is the most tricky one. In fact a lot
> of gifted list members have twisted their brains with this.
>
> I think the calculation of the "sweet spot" for the pivot angle
> is dependant of various factors.
> The self-centering force should be as strong as possible whereas
> the negative trail should be kept small. Also important is the
> weight distribution between front and rear part.
>
> In practice the pivot angle will be somewhere between
> 60 and 70 degrees but the perfect angle needs specific
> evaluation for each individual prototype.
>
> Cheers,
> Jürgen.
>
> ============================================================
>
> This is the Python Mailinglist at freelists.org
>
> Listmaster: Jürgen Mages jmages@xxxxxx
>
> To unsubscribe send an empty mail to
> python-request@xxxxxxxxxxxxx
> with 'unsubscribe' in the subject field.
>
> ============================================================
>
============================================================
This is the Python Mailinglist at freelists.org
Listmaster: Jürgen Mages jmages@xxxxxx
To unsubscribe send an empty mail to
python-request@xxxxxxxxxxxxx
with 'unsubscribe' in the subject field.
============================================================
Other related posts:
