12
$\begingroup$

Yesterday Eliud Kipchoge became the first human to run a marathon distance in under two hours. Part of what allowed him to do it seems to have been that he had pacers running along with him to break the wind. These pacers ran in a strange formation like a "Y:"

running formation

Kipchoge is the white circle. Is there any explanation of how this was arrived at? Was it purely empirical? Is there some physical way to understand why this would be a good formation, from the point of view of fluid dynamics? The inversion of the wedge is very counterintuitive to me.

Improve this question
$\endgroup$
2
  • $\begingroup$ That's very, very tempting to try and simulate it... $\endgroup$
    – tpg2114
    Commented Oct 12, 2019 at 20:58
  • $\begingroup$ @tpg2114: The effect seems to be fairly small at the pace they're running, which is already much faster than I can sprint (4:34 miles). Seems like the thing to do would be to simulate it with bikes. Or maybe make a model and drop it vertically? $\endgroup$
    – user4552
    Commented Oct 12, 2019 at 22:37

1 Answer 1

Reset to default
12
$\begingroup$

Maybe this will help: a picture of the flow around 8 cylinders. This is a viscous 2D flow of an incompressible fluid. The color in the figures corresponds to the magnitude of the flow velocity. A numerical solution is obtained by integrating the Navier-Stokes equations using FEM - see code on community.wolfram.com/groups/-/m/t/1433064

One cylinder is highlighted here (corresponds to the champion position). The drag coefficient of this cylinder is negative! Figure 1

Improve this answer
$\endgroup$
4
  • 2
    $\begingroup$ This looks cool, but could you explain a little about what the plots represent and how they were generated? Does a non-inverted wedge give worse results? $\endgroup$
    – user4552
    Commented Oct 13, 2019 at 1:44
  • $\begingroup$ This is a viscous 2D flow of an incompressible fluid. The color in the figures corresponds to the magnitude of the flow velocity. A numerical solution is obtained by integrating the Navier-Stokes equations using FEM - see code on community.wolfram.com/groups/-/m/t/1433064 $\endgroup$
    – Alex Trounev
    Commented Oct 13, 2019 at 11:16
  • $\begingroup$ Thanks. Maybe you could edit this information into your answer. $\endgroup$
    – user4552
    Commented Oct 13, 2019 at 13:16
  • $\begingroup$ @BenCrowell Thank you. It's done. $\endgroup$
    – Alex Trounev
    Commented Oct 13, 2019 at 13:30

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.