Balancing on bikes

Galt, I’m sorry if you’re put off by the technical jargon. It’s an unintended consequence from 20 years as a military and commercial aviator.

By the way, I tried your experiment. I held my bicycle straight up, then tilted it a little. And sure enough, the wheel turned, just like you said. Then I let go.

Darndest thing, would you believe it fell over? I can’t figure why, shouldn’t the trail have saved it?

Then I picked it up and pushed it. It rolled straight and true for a few feet, slowed a bit, began turning left, and then fell over again. So what dynamics is at work that allows trail to function at speed but diminishes as the bicycle slows? I’m so confused!

If you still don’t get it, let Professor Bloomfield explain it to you:
http://howthingswork.virginia.edu/

As for ruts, they don’t change the laws of physics. In my experience, a rider who looses control in ruts does so because he fails to steer straight down them. He turns the front wheel causing the tire to contact the walls of the rut, slowing the wheel’s rotation and diminishing gyroscopic stability. Not to mention that the knobbie now wants to climb up the side of the rut, wresting control from the rider.

As for the track stand, I have seen track riders come to a complete stop – no back-and-forth – for several seconds. I have seen MTBers remain absolutely motionless for the better part of a minute (fat tires simplify the matter). They might twitch the handlebars, and they might shift their weight about, but the tires remain fixed to the same location on the ground. Oh, and you can find information elsewhere on the Internet stating that on 25 November, 1977, one David Steed kept a bicycle upright and stationary for 9 hours and 15 minutes. Stationary. Motionless. Without rolling. Which sort of refutes the contention that rolling is at all necessary.

As for the link you provided, it refers to “sum angular momentum of the system,” also known as gyroscopic precession. Although precession always occurs with it, it is not synonymous with the basic gyroscopic force.

Considered as a system, when an outside force is applied to any spinning gyroscope, the change manifests itself 90 degrees in the direction of rotation. This is gyroscopic precession, a.k.a. “sum angular momentum.” This is why a bicycle wheel feels like it’s twisting funny if you hold it between your hands and have someone spin it up, then try to turn it. It isn’t turning in the direction in which you pushed, but another 90 degrees around in the direction of rotation. This is also why a top falls over in a gradually decaying spiral when it looses its momentum (before it hits the ground).

If you mount two wheels on the same axle and spin them in oppsite directions, it is the gyroscopic precession that cancels. Precession is directional because the input was directional. Gyroscopic force is_not directional; it acts in a plane and cannot be cancelled.

And, by the way, any definition of torque will include some phrase akin to “twisting force.” Sitting on your bicycle does_not apply torque until you start turning something.

Chronos, you’re looking at my finger and not at the moon. I mentioned coning of rotor blades because it is part and parcel to the physics that is necessary to preventing the rotor from doing a veg-a-matic routine on the fuselage. You can’t create centrifugal force without also creating gyroscopic force; they both are manifestations of inertia being forced to travel in a circle. Like gyroscopic force and gyroscopic precesion, they are joined at the hip.

This is lapsing into a long-winded aerodynamics lecture, but here goes. All winged aircraft owe their stability to a relative equilibrium of aerodynamic forces. Consider a simple airplane flying straight and level. What makes it turn, for instance, is causing one wing to generate more lift than the other. This tilts the vertical component of lift, disrupting the equilibrium that existed when you were straight and level. The weight that had been in direct opposition to lift now needs a counterbalance. Equilibrium is restored when the airplane turns, its now-angular lift component balanced by an equally-angular centrifugal force.

A helicopter has wings too, they just happen to rotate. Although pilots sometimes refer to it as a “rotor disk”, it obviously is not a disk. Neither do these wings create lift in some magical, uniform disk-shape; they obey the same physics as airplane wings and only create lift directly above each individual rotor.

All aircraft – airplanes, gliders, helicopters, gyrocopters, balloons, blimps, dirigibles, zeppelins, hang gliders, parasails and parachutes alike (with the exceptions of military fighters and aerobatic airplanes) – are designed to seek this equilibrium. With the particular exceptions noted, positive dynamic stability is a good thing.

Particularly at a hover a helicopter has little aerodynamic equilibrium apart from that created by the big fan on top. In a manner of speaking, the fuselage is ‘borrowing’ its stability from that generated by the rotor system. In forward flight it gets considerable slipstreaming but even then it still is just tagging along with that oh-so-stable rotor system.

You probably have seen amphibious helicopters, usually with pontoons mounted on the skids. Have you noticed that you never see one shut down the engine(s) while on the water? They can’t. Or start it either. The dynamic forces generated by the slow-turning rotors would cause them to rock so hard they’d tip over.

All helicopters rock about when starting or stopping their rotors, particularly those with 2-bladed rotor systems (Hueys, Cobras, Jet Rangers, etc.). But when the rotors are up to speed they create a gyroscopic force that resists that wobbling and, in resisting, (somewhat) dampens it. And the wobbling is violent enough that it could capsize any amphibious helicopter (including the enormous Chinook).

Land-based helicopters aren’t immune to this transitory imbalance, either. Some helicopters – such as the Hughes/Schweitzer 269 & 300 – suspend their skids from shock absorber-like devices known as an Oleo struts. As their rotors are being engaged, these “shock absorbers” can amplify the rotor-induced rocking in a harmonic phenomenon known as ‘ground resonance.’ If allowed to proceed unchecked, ground resonance will shake the aircraft apart before the rotor reaches operational speed. The pilot’s one consolation is that when the rotor blades come loose, he’ll be safe in the eye of the hurricane.

The pilot relies on that gyroscope for bigger reasons, too. He counts on it being a nice, stable flying disk that he can, through manipulation of aerodynamic forces, cause to change velocity and direction. Fortunately for him, because of its gyroscopic force, it wants to remain horizontal, just like a well-thrown Frisbee®.

All gyroscopes resist change; that is their fundamental property (and the reason they are an essential component in guidance systems). The helicopter pilot uses to his advantage the fact that his rotor system – owing to gyroscopic forces – wants remain horizontal.

If the rotor system did not have its own nominal “steady state,” the whole wretched aircraft would be uncontrollable. And that goes for Chinooks and Kamans and Kamovs with their counter-rotating rotor systems as well. In fact, their rotor systems, like those of all large helicopters, have such a powerful gyroscopic force that the pilot requires hydraulically-boosted controls to be able to budge it.

Sure, a helicopter can fly without gyroscopic force. So can a Frisbee®. Oh, sorry, that’s not called flying, it’s called FALLING.

I apologize for the helicopter tirade but the issue at hand was gyroscopes, and I can think of no more powerful a gyroscope on earth than the spinning rotor system on a CH-53.

Razor blades and ski bikes? You’ve left out roller blades and slalom water skis and ice skates and the Beatles’ skiing chairs. These devices keep the user so close to the ground that no extraordinary dynamics is required; the normal human sense of balance does the trick. The moment arm from the ground to the bottom or the user’s feet is very short. And the moment arm to the user’s center of gravity is, in comparison, extremely long. Not that far from the forces present when you’re walking, actually.

No, because trail only works when the bike is moving. The idea is that you want your base to be below the center of gravity. Suppose the center of gravity on a bike moves so it’s no longer over the base… In the technical jargon, this is called “leaning to one side”. Due to trail, this causes the steering axis to turn. If the bike is standing still, then the turning of the steering axis doesn’t do anything, and the bike just falls over. If, however, the bike is moving forward, then a turned steering axis will cause the bike to move to the side. In fact, on a well-designed (i. e., ridable) bike, the bike will move in the direction of the turn, thus putting the base back underneath the center of gravity.

This would be very interesting, if true. However, there is no such thing as a force “acting in a plane”. The only sense in which a force can “act in a plane” is that the force acts in a line, and the line is in a plane. Every force acts in a line, and every force can be cancelled. Furthermore, the only effect which can be reasonably referred to as “gyroscopic force” is gyroscopic precession, which, as you note, is cancelled by a counterrotating gyroscope.

If you don’t know a subject, there’s no shame in admitting that you don’t know it, and allowing yourself to be educated. Pretending that you know it and making up a bunch of jargon will never get you very far, especially not on a board like this one.

Appeal to authority noted. I’m not put off by technical jargon - I’m put off by the incorrect use of it, especially when it’s used by someone who’s just using jargon for its own sake in hopes of sounding like he knows what he’s talking about.

As I told you before, the slower the bike is moving, the less of an effect the trail has. If you turn the handlebars of a stationary bike, the bike does not start traveling on a curved path, so it does not move the contact point relative to the mass enough to keep it from falling over. When the bike isn’t moving at all, trail doesn’t help any.

You sure are. I’ve explained it twice now.

I see. So this implies that you should be able to ride an unsteerable bike in a straight line without falling to the side. Is that your claim? I’d like to see you try it.

You haven’t been paying attention to the shifting of the mass of their bodies relative to the contact point between the bicycle and the ground, then. Shifting the mass while keeping the contact point stationary accomplishes the same thing as moving the contact point while keeping (most of) the mass stationary.

Funny, that would also imply that torque and torsion are unnecessary, too, but somehow I doubt he was really motionless. See my previous point.

You’re mixing concepts. Gyroscopic precession is the result of the angular momentum of the system. They’re not the same thing. “Precession” does not simply refer to the force being transferred 90 degrees off from where it’s applied – it refers to the effect of this when the force is applied in a consistent way such that circular motion is achieved. This is why the procession rate is measured in Hertz.

Ok, then what’s the effect of this ever-present gyroscopic force which doesn’t get canceled? You’ve already conceded that what you’re referring to as gyroscopic precession gets canceled, so if the remaining “gyroscopic force” doesn’t contribute “gyroscopic precession”, what does it contribute?

Sure it does. All over the place. For example, unless my ass is positioned directly above the seat post, I’m applying a torque to the clamp holding the seat at the proper angle, regardless of whether or not it’s actually moving. Every joint on your frame is likely undergoing some amount of torque.

Your obviously extensive experience with helicopters does not change your misunderstanding of gyroscopic forces.

Galt,

you’ve hit on a great idea! We need to build a bicycle with a fixed front fork - unsteerable and see if CzarNicholas can ride it.

High-wire performers do it. Of course they need a twenty-foot pole to keep their balance - but then they are not royalty.

The distance between your feet to the ground has nothing to do with it (why would it?). In any case, the Razor scooter has the rider’s feet about 2 inches off the ground. On a bicycle, it’s about 4 inches. I fail to see why that makes any difference. And if you looked at the link I provided, you will see that several of the ice bikes in those photos are normal bicycles with the front wheels replaced with runner blades. The geometry is exactly the same as the original bicycles.

Skates and roller blades are different. On those, the user is free to lift a foot and put it down anywhere. Balance is achieved by placing the contact point carefully below the center of gravity. Or more precicely, the average position of the contact points is below the center of gravity. On a bicycle, you can’t lift a wheel and put it down in a different spot. The only way to move the contact point is to turn the handlebar, and let the forward motion move the front wheel to the side. The trail makes this action semi-automatic. So you see why trail works better at higher speeds.

Good news. It turns out David Jones’ article, The Stability of the Bicycle, Physics Today (1970) is available online c/o Joel Fajans, UC Berkeley. Unfortunately, it’s only pdf of a scan, and not a great scan at that, but better than nothing. Easier to read, I found, when printed out (as opposed to trying to read onscreen). Also, be warned it’s a big download (almost 9 MB)

FWIW, in the course of my travels, I came across many articles on the subject, of which these were (for me, at least) the most informative:

An Introduction to Bicycle Geometry and Handling, Karl Anderson (emphasizing side-to-side corrections, but also giving significant weight to trail)

Ask Lou, Physics Central and How Things Work, Louis Bloomfield (slightly different articles, to the same general effect; both emphasize trail, but give weight to side-to-sode corrections)

Exploratoreum (San Francisco) (discussion of Balancing and Steering starts about halfway down the page; continues here) (emphasizing side-to-side corrections)

Forces and Balance in a Bicycle (emphasizing side-to-side corrections)

Balancing a Bike, Herb Weiner (emphasizing trail)

rec.bicycles FAQ, especially 9.35 on gyroscopic forces (debunked); see also 9.15 (discussing, among other things, turns) and 9.16 (trackstands)

How you steer a bicycle, Joel Fajans (good discussion of countersteering); see also Physics & Bicycling

PhysLink.com, Matthew Allen (emphasizing gyroscopic effect, but not discussing Jones)

Hi folks,
this is actually my 2nd post. The first time this was addressed i got so frustrated at Lingle’s response, i had to get in here and post. I see i was ignored the first time, so i thought i’d write again. Oh, i’m not mad or anything, it’s just that it seems everyone seems to take this much farther than it should be. A moving bike balances easier because the rider can readily and precisely move the contact patch from, and to, either side of the center of gravity (CG), causing the bike to fall sideways and turn (i think we all agree it turns by leaning). The only way to move the CG on a stationary bike is to pick it up and move it. Not very easy.

Hey Staubej. It’s not a matter of making it difficult. It is a difficult puzzle. The center-of-gravity theory, which you may note I expressed myself above, explains only part of the story. Mind, I think/agree it’s the part most directly relevant to the original moving/nonmoving bicycle question. For that matter, it was the starting point of Jones’ article. But, as he demonstrated there and others have argued here, there’s a lot more to overall bicycle stability. Which I, at least, find even more interesting than the original question. But, you know, to each his/her own.

BTW, staubej, I did a search for your prior post and couldn’t find it. Only your simultaneous almost-identical post (also referring to an earlier post) in the motorcycle thread. Are you sure it was this board? Or did you perhaps change user name?

Hi PBear,
i too, searched for my other post, w/o any success. It was months ago when this question was first raised and Lingle went into a diatribe about gyroscopic effects. I would agree there is more going on w/ stability, but i was mearly commenting on getting the thing to turn. Thanks for reading!
-john

Folks, here’s what NASA has to say on the matter:

“Any rotating wheel or body tends to stay in its plane of rotation (due to Newton’s laws). That is why a bicycle stops wobbling when you get up speed and why a spinning top stays upright. The spinning bicycle wheel or top gives stability to the system.”

(http://kids.msfc.nasa.gov/Teachers/PastEmail.asp?whichpage=28)

Makes you wonder how they ever managed to put a man on the moon, don’t it.

CzarNick: The NASA statement is correct, but note that it emphasizes speed. Even at moderate to slow speeds, gyroscopism contributes, just not enough to greatly affect the much larger mass of the rider. Nor, of course, does NASA purport to be giving an explanation of all aspects of bicycle stability. It’s merely illuminating (actually, suggesting that teachers illuminate) an abstract physics principle by reference to something to which kids can relate in their everyday experience.

Staubej: No worries. Was just puzzled. Now that I think about it, I think I’ve seen references to lost threads due to a board crash, though I wasn’t here at the time. Maybe that’s the explanation.

Yeah, there were a couple months worth of stuff (everything from Dec. 9th to whenever the board crash was in February) that is lost for good. The first thread on the first bicycle stability column was part of that.

As for the bicycling stuff, I’m just amused to see someone attempting to lecture Chronos on elementary mechanics. It’s especially amusing because the “lecture” come to completely incorrect conclusions.

For example, the track stand is accomplished not by any torsion applied to the frame, but by keeping the contact patches under the center of gravity. Notice how in a track stand, the front wheel is generally turned to the side? That’s because the goal is to be able to move the contact patches from side to side, to counteract the movement of the center of gravity.

Ever tried to ride a bike as slow as possible? You end up steering from side to side rapidly to try to keep the contact patches under you. If I ride my bike in a light snow, I can see the path that the bike follows, and it tends to gently wander a couple inches from side to side. Why is this? Because as I inadvertantly lean a little bit to the side, the front wheel moves slightly towards that side. This puts the contact patches back under me. I can’t say that I’ve ever tried going out to where the trolley tracks run in the middle of the road and riding in the groove, but I doubt that it would be possible.

CzarNicholas, you stated:
>>>As for ruts, they don’t change the laws of physics. In my experience, a rider who looses control in ruts does so because he fails to steer straight down them. He turns the front wheel causing the tire to contact the walls of the rut, slowing the wheel’s rotation and diminishing gyroscopic stability. Not to mention that the knobbie now wants to climb up the side of the rut, wresting control from the rider.>>>

I disagree. Clear your mind and consider this: A bicycle rider who loses control in a rut does so because he has lost control of lateral movement of the bicycle beneath his center of gravity. Gyroscopic effect has nothing at all to do with balance on a bycycle. With both wheels in a rut, I will crash and you will crash. Think about how a unicyclist maintains balance while stationary. It is accomplished by short forward-and-backward movements of the wheel, which affords front-to-back and lateral adjustments of the unicycle beneath the center of gravity. A bicyclist doesn’t have to worry about front-to-back stability because the two tires provide that stability. Look here:

http://boards.straightdope.com/sdmb/showthread.php?threadid=148800

Erm, similar thing with balancing on a skateboard. :-p

CzarNicholas said:

But the reason he turns the wheels slightly is because his weight has shifted slightly to the side, and the wheel turn is an automatic balance adjustment. Thus the sides of the wheels rub the sides of the rut, and cause the aforementioned wipeout. That is precisely the point - the rider has to turn the wheel to maintain the contact patches beneath his c.g.

Also regarding gyroscopic force, you appear to be saying what everyone else is calling gyroscopic force is really gyroscopic procession, and then state there is a completely different force that is gyroscopic force. Your description of gyroscopic force appears to be centripetal force, but then you state that they are not the same thing. So what is gyroscopic force?

the curved part of the tire does not push the bike upwards. it does create a tighter path and smoother area of contact than a square tire though. this helps the rider balance the bike but does not maintain balance itself.

the main balancer on a moving bike is matching speed with the corresponding lean and handle bar turn. those three variables all must be in harmony. gyroscopic force does not help you stay balanced… it is a small force that must be dealt with by the three previously mentioned variables, but does not contribute to an easier way to balance.

the faster the bike goes, the more control a rider has in his small handlebar turns and leans. control meaning, each turn of the handle bars of a given measure will do more work on the system, as speed increases.

starting at equilibrium going 100 mph… if you lean with a force of 5 ft-lbs (which is how any fall will begin… with a lean), you will be able to counteract this force easier, with a handlebar twist, than at 50 mph, because the tire is coming into contact with twice as much road… this extra speed and road contact makes each handlebar twist more meaningful to the bikes’ previous counteractions. each lean/twist produces more G’s the faster and faster you go, which creates a lower center of gravity of the system, (majority of mass (your body) is at top of bike, but the more G’s exerted, the more this “topheaviness” is lessened).

so any lean at a higher speed will result in a more stable, “self-equalizing” system. (when counteracted with either a speed change or handle bar turn obviously)

to summarize: the imperfection that makes you fall to one side or other is a lean. at 0mph the lean is devastating because there is no way to draw center of gravity to bottom point of tire. at 100 mph you are able to counteract the lean with a G force change, drawing the center of gravity back to the bottom of the tire. the easiest way to do this is to speed up so that you are pulling more G’s, so that as far as the system is concerned, your COG is right were it should be.

at 0 mph, if you lean, you will fall unless someone pushes on the top of your head straight into the center of the bike & rider. this would smash the bike into the ground so that you wouldn’t fall, which is same thing as pulling some G’s around a corner.

Fuel, reasonable description, but would be better using realistic speeds. :wink:

Bindlestiff said:

Respectfully, I must disagree. While this technique may be required for motorcycles, bicycles do not require the original push on the handlebars to be in the direction opposite you wish to turn. All that is required to initiate the turn is to lean in the direction you wish to turn and rotate the handlebars to match the lean. While it is possible that using this technique would make the turn easier or sharper, I have not been able to notice this myself. What is important for the turn is to have your weight shift into the turn, so the combination of weight and inertia reacts through the contact patches.

Well, I am not to bright in the physic’s department, I am however the nephew of a ex-pro motorcycle rider. I asked him how the guys on the street bikes lean into turns and not fall over and how they stand the motorcycles back up. He told me that they have to pull it down into the turns and then control how fast the bikes stand up. It explains why you see guys get thrown off the bikes when exiting a turn. If anyone can help explain how and why this works and the connection between this and the initial question