Six things we learned from chatting to Keith Bontrager

Wheels, tyres and more tech-talk with one of the most prolific bike pioneers on the planet

We’ve been indulging our curiosity, finding out more about wheels, tyres and bikes: how they work and where they’re headed. We’ve been learning from Keith Bontrager, a motorcycle racer, frame builder and physics graduate, who turned his hand to making innovative bicycle components in the late 80s, before being snapped up by Trek in 1995.


With scrupulous attention to detail and an ability to think outside the box, he’s been pushing the envelope of bike design ever since, particularly in the interaction of wheels, tyres, and tubeless systems.

BikeRadar recently had the chance to sit down with him, nerd to nerd, over a coffee or two, hoping to draw on his vast knowledge and insights. We weren’t disappointed. These are a few of the things we learned:

1. Forget spoke tension – it’s all about the spoke angle

Bike nerd to bike nerd:

Keith Bontrager: Unless you loosen the spokes until they’re baggy, spoke tension has almost no effect on the lateral stiffness. Spoke bracing angle is the really big deal. When you add more gears, you create more dish; you’re also killing the lateral stiffness and stability of the wheel. When you push the driveside spokes inwards, it just makes the wheel that much harder to make.

Seb Stott: On a standard, non-Boost, rear wheel, the distance between the centre of the wheel and the drive-side flange is just 18mm. When you go to boost, that number increases to 21mm. How significant is that?

KB: Even small changes in that area affect big changes to the lateral stiffness. We used to fight for even a millimetre. We would design a hub based on how close we could allow the spokes to get to the derailleur.

SS: So when Boost was developed, why did they settle on 148 – why not even more than that?

KB: You’re asking the wrong guy! Someone did it, and it worked. If someone like SRAM or whoever does that, then it becomes the universal standard. If somebody like Trek, or Bontrager, or Speciaized says it, then there’s a squabble among the other brands. As long as everybody agrees on a number and it makes sense and does what it’s supposed to, it becomes the number. 150mm might have been better, I don’t know.

2. Wider rims are a no-brainer

KB: The science was never difficult to figure out. Using really narrow rims and wide tyres was never a good idea. Back in the 80s we were basically riding road bikes with knobby tyres – you can save a little weight in the rim by making it really narrow, but you had to run it at 35 or 40psi. Those lightweight priorities were inherited from the road tradition. That, slowly, is vanishing.

3. Carbon wheels don’t have to be fragile

Bontrager gets down to the nitty gritty: bontrager gets down to the nitty gritty

SS: A lot of top racers are reluctant to use carbon rims, especially on the rear. Do you think carbon is inherently more fragile?

KB: The original carbon rims were pretty fragile and if you hit them hard enough they would shatter, but frankly, aluminium’s not awesomely robust – if you hit a rim hard enough to dent it, its lifetime is severely shortened. I’ve broken two carbon rims in cyclocross, and they don’t shatter, for the most part. It sounds like somebody ripping a big branch off a tree, but then they just keep going. Unless you completely blow-out the section, they just snap back into place and there’s a crack. They return to a fairly round state. You can’t dent them.

SS: Carbon rims also tend to be much lighter, so people may be comparing fragility between rims of different weight.

KB: Yes – if you keep the strength and stiffness the same, the carbon rims can be substantially lighter; if you fix weight, the carbon rims can be armour-strong. If I say I want to make an 800g carbon rim, people would look at me like I’m nuts. I mean, why would you want that? Well, you’d want it if you’re a pro downhiller because it would get you to the bottom every single time. That’s a pretty limited market, though.

4. Symmetry is not important

Asymmetrical chainstays on the trek emonda slr 8: asymmetrical chainstays on the trek emonda slr 8

KB: Back in the 80s, I made mountain bikes with asymmetrical chainstays to try to get less dish into the rear wheel (to improve stiffness). When you add more gears, you screw up the dish of the wheel, and you have to do something to compensate for that once everybody realises that having wheels that flop isn’t so much fun… Rims with an offset spoke-bed are another way to fight-off the dish problem.

SS: Do you think there’s a place for asymmetrical chainstays in the future of mountain bike design?

KB: There were so many practical problems trying to explain it to people. People in bike shops freaked out. They would get their gauge out and say, “Your frame’s crooked!” and you’d have to take a half-hour to explain it to them. Then two weeks later, the same thing would happen. Functionally, it works. If you had team bikes with switched-on mechanics it could work, but practically, it’s hard.

SS: Cannondale’s F-Si hardtail uses asymmetrical chainstays to stiffen the rear wheel (by eliminating the dish), but I suppose that would cause uneven load on each dropout – might that be an issue for full-sus bikes?

KB: Motorcycles are like that sometimes. You could do it. Symmetry is just convenient; it’s not structurally important.

5. Your wheels hold you up through a reduction in tension, not with tension itself

Your wheels hold you up through a reduction in tension, not with tension itself, says bontrager:

KB: If you looked at the change in tension when a load is applied to a wheel, the tension in all the spokes away from the contact-patch area doesn’t change. The weight is not hanging from the spokes. The tension is _reduced_ across the contact patch; the wheel is standing on those spokes. The change in tension is the same as the compressive load. It’s hard to get your head around. It’s a pre-stressed structure, which behaves in ways which aren’t obvious.

The same thing’s true of the membrane of the tyre casing which supports the weight of the bike. The tension in the casing changes at the contact patch; everywhere else it stays the same. The forces are supported by the casing tension, and it’s the same negative change which supports the rider’s weight.

6. The bigger your tyres, the lower your pressures should be

KB: That casing tension is what stabilises the tyre in every direction: it supports the tyre against vertical loads and lateral loads too. At the contact patch, where the tyre is pressed flat on the ground, the pressure pushing on the inside of the tyre is balanced by the ground, so the tension drops around it. If you’ve got a tyre at 15psi, you can see ripples where the casing is buckling, as the tension has dropped to about half in that region – I worked that out just the other day!

So if you go from a 2.3in to a 3in tyre, you need to adjust the pressure down, following Laplace’s law, in order to get the same tension. [Laplace’s law states that the casing tension = internal pressure x tyre’s radius.] The way to make the tyres perform the same in terms of structural support is to reduce the pressure in the big tyre so the tension is the same; that’s the fundamental variable. [By this logic, if you’re running 24psi in a 2.3in tyre, you should run 18psi in a 3.0in tyre to get the same casing tension. This is remarkably similar to the pressures I’ve been running when comparing 2.3in tyres to 3in plus tyres in the field – Seb.]


We’ve seen this with road teams – when you switch from a 23c to a 25c tyre, you can’t just run 135Psi without thinking about it. There were times where dropping pressures in the bigger tyres was a good thing from the point of view of traction and suspension, and it doesn’t cost you anything in rolling resistance.