Steering Torque and CG

Two concepts, steering torque and center-of-gravity (or fore/aft weight displacement), have become competing imperatives in bike geometry. Triathletes suffer from too much weight on the front of the bike, because we all lay the top halves of our bodies down on the handlebars. We mitigate this through bike geometry – longer front-centers, shorter chain stays – and by using shorter stems than on our road race bikes.

But there’s a limit to how long our front-centers should be. Front-center is the distance from the bottom bracket spindle to the front wheel axle. The entire range of front-centers in a tri bike size run may only be from 570mm to 660mm, but this is a touchy metric. You can feel a pretty small change.

These changes don’t happen in a vacuum; they’re taken into account along with other elements of steering geometry, like trail, like the steering lever on your bike. But in general, more front-center means a more stable bike until the bike becomes unwieldy and hard to turn. But let’s say your bike’s front-center perfectly suits you. End of story, right?

Yes, 30 years ago. Incrementally less so each decade since then. What’s different? The wheels we ride. Triathletes kept carbon wheel makers alive for a generation, until road riders finally figured out what triathletes knew all those years: deep wheels are faster and their speed trumps the incremental weight increase in most cases. Deep wheels were 40mm, then 60mm, now routinely 80mm and 90mm. More people have them, and the wheels they choose are getting deeper and deeper. This means one has in his or her everyday riding and racing a new data point to contend with: steering torque caused by the wind hitting the front wheel.

How much torque are we talking about? Unfortunately, I don’t know. For about 15 years I’ve been asking those who test bikes in wind tunnels if there is a way they can measure the torque on the fixture in a yaw. So far, no go. This seems to me an important metric, because it would be nice to know what forces the rider is dealing with; and it should be possible to design the torque right back out of the bike, and then test for this. (Now, let me add a caveat. One guy used to test steering torque when he went to the tunnel, and I’ll get to him in a moment.)

Without any knowledge of how the wind affects handling, we are left with rules of competition that don’t make a lot of sense to me. For example, I think there’s much more of an effect on the steering and handling of the bike in a wind if I increase the front wheel’s depth from 40mm to 80mm than there is if I switch the rear wheel from a 40mm rim to a disc. And yet discs are illegal in a number of races while I could show up with my old 100mm-deep Blackwell on the front, no problem.

What I sense on the front is often a sailing effect, like I’m riding the wind like a sailboat in a broad reach, followed by an occasional stall (if I’m riding in a crosswind with a wheel at least 80mm deep). I’m fine on the flats, but it can be a bit scary on descents.

How do you engineer a bike to have less torque affecting the steering? Place surface area behind the steering axis and the contact patch of the tire.

Wind hitting the front wheel from the side turns that wheel about the steering axis in the direction of the wind as you ride. What I think happens (I’m not certain about this) is a confluence of forces that conspire to give you handling problems with these wheels.

First, wind hitting in front of the steering axis and in front of the tire’s contact patch pushes the wheel one way, and the wind behind the steering axis and the contact patch pushes the wheel the other way. Second, the wind behind the steering axis is dirtier and doesn’t hit the back of the wheel with the same clean, unobstructed forces as it hits the front of the wheel. Further, the front of the wheel is cantilevered well in front of the steering axis and that extra lever amplifies the effect of the wind.

And finally, if that weren’t enough, you have front hydration systems built with surface area and fun and funky super bike front ends, all of which sit in front of the steering axis.

In the diagram above the area in green is what I contend (and it’s just my guess) is the area forcing the wheel to turn in the direction of the wind. That is, when the wind hits the green area the wheel is pushed in the direction of the wind. Air hitting behind the steering axis (which is in line with the bicycle steer column) forces the wheel back in the other direction, but there’s less surface area back there. The light brown represents all the stuff on your bike that functions well (like aero front hydration systems) but adds to the steering torque you must contend with and overcome when you ride.

How would you counteract this? By placing surface area behind the steering axis. This extra surface area must be attached to the front half of the bike, that is, to something that rotates about the steering axis. Like the fork, for example. If your fork had a trailing edge roughly approximating the violet area on the diagram above, that might reduce the torque applied about the steering axis.

Is this likely to be made? I don’t know. Would it be more or less aerodynamic than how the bike is now? I don’t know. Would it be legal? Don’t know.

What I do suspect is that front-centers too long present a yet longer lever, generating yet more torque, when you’re trying to steer the bike in a wind with deep wheels. For this reason, and while I like enough front-center to make the bike handle nicely, and give decent weight displacement, deep wheels argue against too much front-center. I’m 6’2” tall and I ride reasonably steep on my tri bike and I really like bikes with about 630mm or 635mm of front-center. But today's bikes are all coalescing around 625mm of front-center, when these bikes are made in a height that works for me, and while I’m not all that red hot happy about that loss of front-center, one by-product is that a deeper wheel seems a bit easier to control.

One manifestation of this are bikes that are too long for you, where you must draw back the armrests and extensions to make them fit. This probably results in a front wheel well in front of you, and the extra protrusion of that front wheel axle means you may need to wrestle with that bike a bit in a crosswind if you're riding a deep wheel.

That one guy to which I alluded above who used to test for steering torque? John Cobb. He wrote about this on Slowtwitch 16 years ago, and I’ve republished the article today. I went back and reread it in preparation for this article and found it fascinating and prescient.

In a prior article I scribbled a little guide for determining front-center scaled up and down according to size. If you measure the run from the bottom bracket forward to the place where you hold the aerobars (the joint where your index finger meets the palm of your hand) you'd have as follows:

BB-to-handhold => FC
86cm => 660mm
84cm => 650mm
82cm => 640mm
80cm => 630mm
78cm => 620mm
76cm => 610mm
74cm => 600mm
72cm => 590mm
70cm => 580mm
68cm => 570mm
66cm => 560mm
64cm => 550mm