How Bikes Steer Themselves | Hacker News Reader

How Bikes Steer Themselves

sciencefriday.com

38 pointscelias14 comments

http://en.wikipedia.org/wiki/Bicycle_and_motorcycle_dynamics has all you need to know on the subject.

I'm not sure I agree with the section on countersteering. It says that "out-tracking", the fact that you steer the wheel out from under the bike, is more important than than gyroscopic precession. For motorcycles on highway speeds, I think this is incorrect. If you've ridden a motorcycle at high speed, you know that the handlebars are basically like buried in solid concrete. With normal steering force, I don't think I deflect the handlebars by anything noticeable. The key is that at high speed, the mere application of torque on a spinning wheel, will make it precess, ie turn around the orthogonal axis. I'm convinced this is far more important (but I'm too lazy to do the numbers now).

A motorcycle can be turned at speed two ways. Out-tracking is only one of those ways. (Leaning is the other)

However, it is the way that lets you take a deep set most quickly, and can help a lot with tight lines.

You may think that you are not out-tracking when you turn at freeway speeds, but remember the amount of necessary out-track decreases as speed increases.

You may think that you are not out-tracking when you turn at freeway speeds, but remember the amount of necessary out-track decreases as speed increases.

If you think that out-tracking is responsible for entering turns at speed, there's an experiment to test this: If the bike leaned by moving the wheels out from under it, the bike should pivot around the center of mass. If you actually look at a bunch of road racers in the twisties, it's abundantly clear that the bikes pivot almost completely around the contact point, not around the center of mass. (Unlike bicycle racers which really do out-track.) Moving the COM without moving the contact point can not be done through wheel forces on the pavement, because they act at the contact point. It's impossible for those forces to lean the bike and move the COM is the same direction. The only thing that can accomplish such a motion is the precession resulting from applying the handlebar torque to the front wheel.

I'm also not sure how this is going to help someone build a better bike... just because it's possible doesn't mean it's a good idea. bikes with steep head tube angles are jerky.

Rake is your friend.

This became patently obvious when I rode motorcycles that ran the whole gamut of rake angles. Even more obvious when I then remounted a bicycle, which has drastically less rake than even the most aggressive motorcycle.

This is a great example of not having to understand a phenomena in order to exploit. Of course, understanding will let us take it even further.

As best I can tell, it's a mix of forces.

Gyroscopic action helps keep the bike upright by slowing falls. (at 100mph on a motorcycle, you must apply significant force to lean over quickly)

Rake (trail) makes it less likely the front wheel will be deflected by small bumps, and I suspect it also helps settle the bike back into neutral position. In other words, decreases the number of significant self-steering corrections that must be made.

Then, some third force does the wheel turning. I personally suspect it is tied to the way the front wheel falls to the right if you suspend the bike and lean it right.

Note: Honestly, from the diagram at 2:03 it looks like they did not eliminate gyroscopic action and trail. They merely greatly reduce gyroscopic action, reduce rake, and invert trail.

To turn/lean a bike at 100MPH you have to push the handlebars opposite of the direction you want to turn. If you need to swerve right, lean right and push the handlebars left. Not knowing this is probably related to many loss of control accidents among novice riders (speculation).

Of course. My point is, the gyroscopic forces of a pair of 20lb wheels spinning at nearly 1,000rpm are not trivial. It takes a noticeable amount of force to make the bike respond quickly.

Do people really not know a bike will steer itself? it's been called ghost riding for a long time... even before that ridiculous song.

know that a bike will steer itself != know how a bike steers itself. It's a pretty nontrivial coupling of geometry with conservation of momentum and angular momentum.

at 0:31 the guy says "I always have to tell people that bikes do stay up by themselves." I find that surprising... that people don't know.

Reply on news.ycombinator.com