WE WERE asked by New 杏吧原创 to investigate the elastic properties of the soles of a selection of running shoes, paying special attention to 鈥榚nergy return鈥� (see Table 1). We decided that our tests should be simple, so that other suitably equipped laboratories could easily repeat them, and resemble the patterns of force that would act on shoes on a runner鈥檚 feet.
Most runners hit the ground with the heel first, creating a brief peak in 鈥榞round force鈥� on impact. But the centre of pressure of this force moves rapidly to the ball of the foot, reaching its main peak, and remains there until the foot leaves the ground. At this stage, the forepart of the sole is compressed and then rebounds, as the rest of the body decelerates and reaccelerates. The impact peak is typically 1.5 to 2 times body weight, for runners wearing shoes and running across a rigid force plate, and the main peak 2.5 to 3 times body weight. A typical man, of mass 70 kilograms, has a weight of 700 newtons. With this information in mind, we decided to apply peak forces of 1500 newtons to the heel, and 2000 newtons to the forepart of the sole.
Advertisement
The foot remains on the ground between 0.2 and 0.3 seconds depending on the speed of the runner. So we performed the tests at a frequency of 4 hertz, making the force rise from zero to its peak and return to zero in 0.25 seconds. The heel remains on the ground for much less time than this, so we also performed tests on it at a higher frequency, 11 hertz.
Our results for heels show substantial differences in the extent to which different shoes deformed under the pressure (see Table 2). Deformation under the load of 1500 newtons ranges from 7 to 15 millimetres, at 4 hertz. The more compliant shoes 鈥� the ones that deform more 鈥� require more energy to deform them. At a frequency of 4 hertz, energy return ranges from 55 per cent to 69 per cent. The tests at 11 hertz consistently give similar, but slightly smaller, deformations and energy returns.
The tests on the forepart of the sole showed a similar variation in the percentage of energy returned, but much less variation in deformation (see Table 3). All the shoes deformed between 9 and 12 millimetres under the peak load of 2000 newtons.
But high return of energy from the heels is unlikely to matter, because the heel recoils too early in the step to be useful to the runner. The kinetic energy of the foot is lost when the heel hits the ground and is probably inevitably dissipated. A high return of energy from the heel may simply keep the vibrations following impact going for longer.
The quality to look for in a shoe鈥檚 heel is probably high compliance (or peak deformation), which will reduce the forces of impact. Looking at the same thing in a different way, the heel should be able to absorb the foot鈥檚 kinetic energy without developing large forces, so large values in 鈥榩eak deformation鈥� and 鈥榳ork of deformation鈥� are probably good.
High energy return in the forepart of the sole does seem potentially important (see Table 3). But the figures from our tests for the percentage of energy returned still do not tell us how much energy a shoe returns. For that, we need to know how much energy was stored in the first place. The higher the compliance (or peak deformation) and the work of deformation, the more energy the sole stores as it is compressed. The higher the energy return, the more of that energy is recovered in the elastic recoil. Looking at our results in this way suggests that Asics and Avia, for instance, return more energy than Hi-Tec or Turntec, with the other shoes falling somewhere in between.
But you should treat our results with caution, for a number of reasons. We tested only one individual shoe of each brand. Furthermore, any apparent discrepancies between our results and the claims of manufacturers may be due to different methods of testing. To complicate matters, experts may disagree about what makes a running shoe good. Our tests give no information about many important properties, such as stability, comfort and price. Most importantly, perhaps, small differences in the energy returned may not count for much. The work of the deforming the shoe and the energy returned in the elastic recoil amount to less than 10 per cent of the total energy (kinetic plus potential) lost by the body and regained during each step, which amounts to about 100 joules.
R. McNeill Alexander FRS is professor and Michael Bennett research fellow in the department of pure and applied biology at the University of Leeds.