Re: Instability and Speed Wobble: Bikefanclub.com

A colleague of mine mentioned that on the last AIDS LifeCycle Ride (SF -> LA), there were more accidents than in years past -- one especially nasty one that was apparently the result of speed wobbles, or wheel shimmy. I have never read anything online about this, so I was surprised to find out that there were so many competing theories when I got back to the office and looked.

Before searching online, however, I simply said to her, "It's all physics. Remember that movie of the bridge up in Washington that we all saw in elementary school? Like that bridge, you plus your bike have a very deterministic and discreet set of resonant frequencies, depending on how you are seated, if you are pedaling, hands on or off, etc, etc. The rotation of the wheels (probably only the front really), in combination with your exact frame geometry and handlebar length and weight give rise to the minute oscillations that resonate with those discreet frequencies I just mentioned. If you get the wobbles, loosen up your grip (or put hands back on the bars, if they're off), clamp the top tube in between your legs (thighs, calves, whatever), and either slow down or speed up." [Yes, I actually talk like that. And I sound like even *more* of a pompous ass than you'd guess from that .-]

This is all based on my own experiences on 4 different road bikes and 3 different mountain bikes over the past 30 years of riding (Yes, I'm an old fart now), and on my Master's education in Physics.

If anyone is interested, here's what I experienced:

1.) 1981 Bob Jackson (19" steel frame/fork, Ambrosio tubulars, 32 h): perhaps the finest hand-made bike I've ever ridden. I had speed wobbles between 45-49 mph, but only at apparently random times. Changes to the bike (new tires, lowering/raising the seat, etc) did nothing to affect the range in which wobbles occurred. The headset is a 1982 Campy Super Record that has been checked by many mechanics. It's definitely not that part. On all bikes that I have had shimmies, I have found that when I keep my elbows locked, the wobbles are MUCH worse; keeping them loose significantly reduced the effect.

2.) 1983 Peugeot (steel, Mavic clinchers, 32 h): pretty much the same here, except the wobbles occurred at a bit lower speeds (40-44 mph).

3.) 2002 Raleigh Team (aluminum frame, carbon forks and seatstays (both Columbus Carve), Campy Proton clicher wheelset): this bike is a bit taller than my others (it's a 59cm), and the wobbles I have experienced are much more pronounced on this bike than any of my others. Also, I can predict each and every time it will occur (well, almost every time). It always happens when I hit 43-47 mph (depending on the road condition: curving, straight, bumps, smooth, etc). It's gotten so that I don't even freak out at all anymore... I simply clamp the top tube between my thighs and the wobble's frequency and amplitude are diminished pretty much to zero right away. I have also found that simply cranking up to bit higher speed does the same. Like a hockey puck traveling thru rough ice speeds up when it returns to smooth ice, the shimmies just disappeared when I got back above 49 mph or so.

[Note: I don't ride this bike anymore .-]

4.) 2006 Ridley Damocles (carbon, carbon, carbon ;-), Fulcrum Racing 1 clincher wheelset): yowza! this bike is insane! I have practically tried to induce wobbles on this thing and I can't get it to budge an inch. It descends like an arrow fired straight down from a building rooftop. Anyway, I've yet to experience any shimmies on this bad boy. (Of course, I've yet to break 60 mph... .-)

5.) Mountain bikes -- 1998 Giant MCM 980 (carbon front, aluminum rear tri,d/s), 1999 KHS (steel, hardtail), 1991 Cadex 3 (carbon w/ aluminum lugs, hardtail): never got any wobbles on any of these steeds. Lucky me!

So, what does all that mean? Basically, to me it means that wheel wobbles, or shimmies, are basically the result of hitting the resonant frequency of the system, in this case the rider + the bike. This doesn't mean that uneven (non-symmetrical) forks do not contribute to it in some way, or that rider position and action (ie, locked elbows and knees) are the sole cause -- it's all these things and more.

From the physical point of view Mr Brandt is most accurate in describing the action: "Shimmy is caused by the gyroscopic force of the front wheel whose tilt is roughly at right angles to the steering axis, making the wheel steer to the left when it leans to the left." In this case, replace "shimmy" with "torque", and you're basically back to Physics 101. Very straightforward stuff. You spin the wheel, it wants to pull left. Lean left, and it wants to pull even more.

But the gyroscopic motion of the front wheel is only one of the variables involved: the height of the bike, the rigidity of the wheels (their tendency to resist deflection), the elasticity of the frame material(s), the center of gravity of the system, the list goes on and on (and on...). Just as the small oscillations of the wind across the bridge in WA caused massive resonant swings in the structure, tiny oscillations in the front wheel become vastly magnified when they resonate with the rest of the system.

To truly model the system accurately would be a nice doctoral thesis, I think. No kidding: there have been several doctoral theses on the physics of skipping stones across water, which, it turns out, is far more complicated that one would imagine just watching those stones fly across the pond. I can imagine that this would be of an equal or greater degree of complexity.

In the final analysis, what makes me convinced that the effect is mostly a resonant frequency issue is that when I get out of the "resonant band", either by slowing down (carefully) or speeding up (much more fun, when possible), they dissipate. One thing I NEVER do is maintain the same speed and attempt to "control" or "muscle" the wobbles away. Don't fight physics, use it .-)

Of course, after all that, you can just look at any of the many motorcycle crash videos on YouTube: it's pretty clear (to me at least), that asymmetrical forks ain't what's doing it. Or, to really feel it, jump on a skateboard and hit the nearest hill with your knees locked. When you begin to wobble, relax your knees: look ma! no wobble. (By relaxing your knees, you effectively change the resonant frequency of the system; at least until you speed up enough to hit the new, higher frequency about 100 yds and 15 mph+ later .-)

But, then, these are my own experiences and conclusions; YMMV.

Enjoy!

Post Extras