That sounds interesting, I didn't know your program was testing self-stability. Have you also written some associated documentation on the underlying physics it simulates? I.e. which eqs. you implement? I didn't see a 'user manual' of the program in the archive. (Not sure how to interpret correctly the inputs and outputs of your program).
18 parameters is a space too large to scan. But if you restrict the parameter space to only a couple at a time (e.g. exploring 2 parameters around a working configuration, for example), you could represent regions of self-stability beautifully (phase space diagrams in terms of 2D contour plots). If you then take the list of pythons in Jürgen's website and show where they are in these phase-space diagrams, we would see how much of that space has been already explored. That may also provide experimental verification!
I am not sure what you mean with 'pivot torsional K': is that a torque you apply at the pivot from the rear part to the front part (in a turn) (somehow simulating a rider)?
Thanks anyways to provide your program! Regards, Pascal On Sun, 17 Apr 2011, RayS wrote:
Interesting thread - the self-stability program I wrote http://rjs.org/Python/FrameGeometry.zip shows the effect the researchers wrote of pretty clearly. It does also have input for rim/tire mass (mouse over the boxes for explanation). If a python's: - front mass is at least 45cm in front of the pivot line - rear mass is <24cm to rear axle - trail <11cm -wheel mass is low it is nearly neutral. All these things have been covered on the list one way or another, except wheel mass. Panniers off the back, shifting weight backward, reduces self-centering force on the pivot. Weight in the very front can increase some desired flop, but is a sensitive factor and makes slow riding more tiresome. Trail should generally be minimized. Importantly, the dynamics are very sensitive to wheel mass; heavy wheels, especially the front, eliminate any chance of stability! If you think about it, gyroscopic effect prevents the front wheel from responding to lean, which is what gives bikes stability; it actually turns the forks the opposite way when leaned. Remember all this has to do with self-stability, and not necessarily how a python "feels" to ride using leg steer. Note that in the attached I set the pivot torsional K to -7; a small opposite force like your hips make when riding the python which counters the self-center effect. It then has a wide range of stable speed. The further mass is from the rear axle the more -K is required, and the lower it is the less stable. It would also be interesting to let the code grind through all reasonable combinations of the 18 variables used and see where the islands of stability are. Ray At 08:33 PM 4/16/2011, Vi Vuong wrote: How about a really heavy front tire? I have been wanting to use car tires like those cool choppers, to build an exercise machine (slow, heavy, abs workout...) , but maybe it can be used for the riderless experiment also. The 30-lb spare tire should have enough momentum to overcome BB flopping; see attached. Rear wheel probably should be small, say ~16in, with added weights if necessary. Also, I am not sure if pivot should be variable or just fixed at 60 deg. Vi From: "pybuen@xxxxxxxxx" <pybuen@xxxxxxxxx> To: python@xxxxxxxxxxxxx Sent: Sat, April 16, 2011 11:21:55 AM Subject: [python] Re: A Bicycle Can Be Self-Stable Without Gyroscopic or Caster Effects I was proposing first a much simpler experiment: putting various weights on the seat and launching the python at various speeds riderless (i.e. control-less) (do I need to put a disclosure notice? You guessed it). This experiment works well with upright riderless bicycles, even without weight. From what we can see in the video that Vi posted, the python self-steers to the left with a left lean at small weights, which would, it seems, counteract the fall, but the opposite self-steering effect is seen at heavier weights (it steers to the right with a left lean). Is this latter behaviour the source for having to relearn how to ride with a Python? Redistributing the weight more on the front wheel with a pole could be interesting too... However, this would also be quite different from putting weight directly on the front wheel (i.e. playing with new sitting positions as DirkS suggested!). It also depends very much if this weight is ahead of the wheel-road contact point or not. Anybody went for a bicycle trip with loaded front wheel? Are you saying DirkS that this would be unridable? Why do you think the CoM should be as close as possible to "the front"? In fact, a substantial redistribution of the weights is effected when riding uphill (rear weight) or downhill (front weight). Is one way more stable to ride (not speaking of fear of speed)? One could also experiment unloading: attaching Helium balloons to various parts of python... :) A lot of fun and spectacle for the kids... pricey perhaps? Mh. Still less than the moon... (some may remember E.T...) A cheaper alternative would be to ride underwater :) In any case, riderless behaviour is not strictly equal to rideability - we can ride upright bicycles at very low speed too, where we know they would fall without our skillful presence. Pascal On Sat, 16 Apr 2011, dirk@xxxxxxxxxx wrote: > Hi, > > > now that explains this forward pointing rod with an extra mass attached to it > :-) > ?? how would we do this in real life? > Sit in front of the front wheel, while still being attached to the rear wheel? > Or have an extra mass greater than the drivers weight pointing forward? > scary :-) > ?? The only practical clue i can derive is that among other things one should have > more weight on the rear part than on the front and aim for a centre of mass as > close to the front as possible. > > ?? Greetings, > DirkS > > ?? ?? > "Jürgen Mages" <jmages@xxxxxx> hat am 15. April 2011 um 13:32 geschrieben: > >> Thanks Christian. The clue is: if the rear part's center of gravity is >> shifted far enough over the front part, then the negative trail bike is >> inherently stable. >> >> Cheers, >> Jürgen. >> >> On 15.04.2011 12:11, Christian Andersen wrote: >>> Hi folks >>> >>> In Science there is an article about bikes, that could be interesting >>> for those of you guys, who have sufficient knowledge (unlike me). >>> Maybe it is worth to have a look at to improve the pythonconcept >>> regarding stability. >>> >>> http://www.sciencemag.org/content/332/6027/339.abstract >>> >>> Science 15 April 2011: Vol. 332 no. 6027 pp. 339-342 DOI: >>> 10.1126/science.1201959 >>> >>> * Report >>> >>> A Bicycle Can Be Self-Stable Without Gyroscopic or Caster Effects >>> >>> 1. J. D. G. Kooijman1, 2. J. P. Meijaard2, 3. Jim M. Papadopoulos3, >>> 4. Andy Ruina4,*, and 5. A. L. Schwab1 >>> >>> Abstract >>> >>> A riderless bicycle can automatically steer itself so as to recover >>> from falls. The common view is that this self-steering is caused by >>> gyroscopic precession of the front wheel, or by the wheel contact >>> trailing like a caster behind the steer axis. We show that neither >>> effect is necessary for self-stability. Using linearized stability >>> calculations as a guide, we built a bicycle with extra >>> counter-rotating wheels (canceling the wheel spin angular momentum) >>> and with its front-wheel ground-contact forward of the steer axis >>> (making the trailing distance negative). When laterally disturbed >>> from rolling straight, this bicycle automatically recovers to upright >>> travel. Our results show that various design variables, like the >>> front mass location and the steer axis tilt, contribute to stability >>> in complex interacting ways. >>> >>> >>> greez, >>> >>> christian a wannabe pythonrider >> >> ============================================================ >> >> This is the Python Mailinglist >> >> //www.freelists.org/list/python >> >> Listmaster: Jürgen Mages jmages@xxxxxx >> >> To unsubscribe send an empty mail to >> python-request@xxxxxxxxxxxxx >> with 'unsubscribe' in the subject field. >> >> ============================================================ >> ?? > http://dirk.steuwer.de
