I don’t think that you could maintain balance on something like that for any period of time whatsoever, and the direction of your fall would be determined at the time that you last had contact with the ground. As soon as your feet left the ground, or whatever was holding you upright went away, you would immediately start leaning in one direction or the other, and not be able to do anything at all to stop or reverse it. Once the center of gravity is on one side of the contact points or the other, fate is sealed, and there is no way to move the center of gravity to the other side.
Think of it as being over a single point. If your center of mass is not directly over that point, you will start to fall in that direction. What could you do to change that orientation. Is there any solely inertial move that you could make that would put your center of mass back over the center? It’s the same thing with the two contact points. You are now limited in direction to fall from any direction to only two, but you still have the problem that you will fall in the direction that the center of gravity demands, and there is nothing you can do about that (without putting your feet down, or turning the wheel[impossible as it is fixed in this example]).With a big enough fan, you might be able to get enough air resistance to stop or reverse, or even maintain balance, but that is using outside forces, not just torque.
Right. The bicycle+rider is an assembly that is free to pivot around the tire’s contact patches. There’s no way to apply torque to the assembly. One could shift the weight distribution, and temporarily apply a lateral force to the pivot, but it will be followed by a equal and opposite force as soon as it stops moving. The center of mass of the assembly stays in the same place (same angle relative to the pivot).
Note that there are “ski bikes” out there that are easily controlled without gyroscopic forces. They look like typical bikes but have tandem skis instead of tandem wheels. They ride pretty much like bicycles, and if you can ride a bicycle you can ride a ski bike.
Note also that the Wright brothers knew about counter steering. Also, Kieth Code built a “No BS” (body steer) motorcycle. It’s a motorcycle with two sets of handlebars, the original set and another that is attached to the chassis, not the front wheel. The challenge he issues is for anyone to ride the bike using body steering instead of counter steering. You can go to this URL for one article about the No BS bike. http://superbikeschool.com/about-us/machinery/no-b-s-machine/
Lastly, I haven’t actually tried this out yet, but I don’t believe counter rotating gyroscopes “cancel” each other’s gyroscopic effect. My experiment setup will be a triangle of wood with two bicycle wheels, one on either side, solidly attached at the apex. Spinning one wheel adequately fast should hold the device vertical. Spinning the other wheel in the opposite direction will… Well, I’ll have to try it out to see.
I asserted that it was impossible to not apply a countersteering force, not necessarily a countersteering angle–although in practice, with freely turning handlebars, the latter will basically always happen. If you locked the handlebars but installed a force sensor, you would detect a brief torque opposite to the direction of lean (due to trail).
Of course the center of mass doesn’t stay in the same place. Let’s simplify the model down to a massive assembly with a single hinge in the middle and a contact point at the bottom (of course, a bicyclist has many hinges, but one is enough). If you were on a frictionless surface and bent the hinge, the bottom contact point would slide to one side. If instead you’re on a surface with friction, the reason the contact point doesn’t slide is because the friction is exerting a sideways force on the assembly. And because the assembly has a net sideways force on it, the center of mass will move horizontally.
And this is what I meant when I said that most experiments which purport to prove countersteering do nothing of the sort. I agree that I could not, on such a bike, make a left turn. But that’s because I can’t turn the handlebars left. It’s not because I can’t turn them right. Code successfully proved that you need to use the handlebars to steer. He didn’t prove how you need to use them.
You’re essentially proposing a third mechanism. The no-BS bike demonstrates that you can’t get the bike tipped over with weight shift alone. And you deny that centrifugal force is used to tip the bike (since that’s the actual force at work with countersteering). What else is there?
Or, maybe, you deny that tipping the bike at all is necessary. In which case I’d wonder if you ever removed the training wheels from yours :).
Ok, so at the least you acknowledge the earlier critique of your experiment, which is that despite your efforts to only push the handlebars in one direction, you’re subconsciously shifting your body mass.
Still–since turning the handlebars in one direction could only cause a centrifugal force in the opposite direction, you’re suggesting that shifting your weight overcomes both this force and then some additional force to tip the bike further. But the no-BS bike demonstrates (for motorcycles, at least) that even inducing the tip is difficult to impossible. If it’s not possible for even a portion of the force, then it’s certainly not possible for the whole amount.
I’ll also note that the no-BS bike is a motorcycle, which is heavier than a bicycle. Countersteering might be necessary on a motorcycle, and I can’t disprove that, because I don’t ride one of those. I’m just saying that the no-BS bike doesn’t prove it.
I agree that you can’t ignore weight–I’ve only ridden motorcycles enough to learn that the effect of shifting your weight around decreases quite rapidly as the bike mass goes up. I barely rounded a cul-de-sac in my dad’s Harley, largely due to me being uncomfortable with the required countersteer. Much less is needed on a bicycle or even a lighter motorcycle, and it can be more easily induced by weight shift.
Because all most people know about gyroscopes is that they act weird. Without knowing the specific details of how and why they act weird, one wouldn’t know that it’s all vectors, which therefore all cancel out.
Two counter-rotating gyroscopes will more or less cancel each other out (since both gyroscopes can’t be in exactly the same place, they won’t exactly cancel, but close enough).
But you don’t really need to do this experiment for bicycles: a little googling will find that people have already built bicycles with an extra, counter-rotating wheel. They’re just as easy to ride as regular bicycles, showing that the gyroscopic effect is negligible for a bicycle+rider.
Not exactly. If you shift the cg of the bike/you pair to the side of the forward path (i.e. lean), the geometry of the bike (i.e. trail in the front wheel) causes the front wheel to turn into the lean. Turning the front wheel is what controls where you go. Handlebars give the rider away to do that directly.
You don’t even need to ride a bike to see this behavior. What happens when you put the kickstand down? The bike leans toward the kickstand side, and the front wheel turns.
I think the point is that he doesn’t believe the gyroscopic forces are canceled on those bikes. So of course they are easy to ride.
Not exactly. When the bike leans (to the left for example) the front wheel “wants” to turn to the left, but you don’t let it. If you were to let the front wheel turn to the left, it would straighten up the bike. When you are in a constant radius turn, the front wheel is very nearly neutral. (straight) The actual turning force comes from the shape of the tire and the fact that the contact patch is now up on the side of the tire.
Turning the bars controls the lean. The amount of lean determines where you go.
The above doesn’t really apply at parking lot speeds where the bike is essentially being balanced upright at all times. At low speeds the bike is generally steered where you want to go with small changes in the steering to keep the bike balanced.
Rotary encoder has arrived and I have it hooked up. See my writeup here.
No useful data yet. Roads around here are too rough for useful results. Not sure I’ll have time to find a big empty parting lot tomorrow, so you might have to wait a week for more, but you might find the rest of the writeup interesting in the meantime.
To cure the drift and resolution problem at the same time, McMaster has plastic gears you can buy, if you don’t mind spending some money. Get a large one, bore it out if necessary, and install it on your bike’s handlebar stem (you’ll need to remove the stem to do this); then install a smaller gear on the encoder. Now you’ll have no drift, and far more pulses per degree of steer angle. The noise issue of steering/balance corrections will still be present if you’re riding on rough roads, but if you can find some smooth pavement, I think you may get some good results.