Heriot-Watt University student James Carchrie has devised a system of dynamic wheel weighting which aims to store and release energy in a manner that helps maintain speed during momentary drops in athletic performance.
If you’ve ever ridden with a pair of properly heavy wheels, you’ll be familiar with the flywheel effect they produce.
Heavy wheels are hard to accelerate up to speed but once they’re spinning, they have a tendency to want to keep spinning.
Citing the case of Ondřej Sosenka, who in 2005 set the hour record on a bike with a weighted rear wheel, Carchrie proposes that a system of sprung moving weights mounted on the spokes could be used to exploit the flywheel effect at higher speeds.
Carchrie’s wheel is a have-your-cake-and-eat-it design — at low speeds the weights stay near the hub, meaning that the moment of inertia is lower.
As speed increases, the weights move outwards, getting closer to the rim. This increases the moment of inertia, and hence the flywheel effect.
At a certain speed, magnets hold the weights in their outer position, meaning that small dips in speed won’t cause them to move inwards; the rider has to slow considerably before the springs overcome the magnets, and the weights move in again.
The size of the weights and the strength of the springs and magnets determine the precise behaviour of the system.
Proving the concept
Carchrie constructed a test rig attached to a turbo trainer to demonstrate the principle, and with seven 475g weights attached to the spokes he was able to simulate a system which achieved the desired flywheel effect at around 40km/h, with the magnets being pulled back to the inner position only when the speed dropped to 34km/h.
There is of course no such thing as a free lunch in physics. The system of weights doesn’t give you free energy, it just stores and releases the energy the rider puts out.
In addition, you do still have to carry that extra weight around, which would itself increase the total energy expended.
Nevertheless, it’s an intriguing concept and it’s certainly possible to imagine applications for it, particularly in a flat time-trial type situation where weight is not the biggest concern.
If you're interested in a fuller explanation of the dynamically weighted wheel concept, you can read Carchrie's full dissertation on the Heriott-Watt website.
