Synchronization from dissipation?

[Demystifier]

TL;DR

What is a simple explanation of the effect shown in the attached video?

The video shows an interesting effect of synchronization of oscillators. The most confusing property of this phenomenon is that it is not time-inversion invariant; if the oscillators started in the synchronized state of motion, they would not end up in the unsynchronized state. This means that dissipation also must play a role in this effect. But what exactly the role of dissipation is? It's not clear to me. Allegedly the effect can be explained by the Kuramoto model https://en.wikipedia.org/wiki/Kuramoto_model but I don't understand that. Can someone give a relatively simple explanation of the effect?

 

[Lord Jestocost]

"The results are well described by a simple model. The metronomes are described as van der Pol oscillators¹⁷ and the coupling between the metronomes comes from the undamped motion of the base."

From "Synchronization of metronomes" by James Pantaleone
[PDF]
Synchronization of metronomes - Department of Mathematics

 

[DennisN]

I'm not very experienced when it comes to the physics of these kinds of dynamics, but my thought after seeing the video was that I thought it may have something to do with resonance (resonant frequency) and the page you posted mentioned intrinsic natural frequency which is related to resonant frequency. But I have no thoughts to share at the moment about dissipation.

A very cool video, thanks for posting!

Sidenote: Another cool (very) dynamical system is the double pendulum, https://en.m.wikipedia.org/wiki/Double_pendulum, which has sort of opposite dynamics (always chaotic), which you may be familiar with.