Why pendulums swing in harmony – new theory

Posted: July 27, 2015 by oldbrew in Measurement, research
Perfect harmony? [image credit: homedit]

Perfect harmony? [image credit: homedit]

From the believe-it-or-not file, Phys.org reports a possible solution to an old puzzle:

Almost 350 years ago, Dutch inventor and scientist Christiaan Huygens observed that two pendulum clocks hanging from a wall would synchronise their swing over time.

What causes the phenomenon has led to much scientific head-scratching over the centuries, but no consensus to date.

‘But now’ – as Tomorrow’s World presenters used to say…

On Thursday, a study in the Nature journal Scientific Reports proposed a solution: the pendulums transfer energy to one another through sound pulses.

A pair of Portuguese scientists hypothesised that these pulses might move from clock to clock, perturbing the swing of the pendulums and eventually causing them to synchronise.

They developed a complex mathematical model before conducting experiments with a pair of clocks attached to a rail fixed to a wall.

The theoretical predictions and simulation matched, they found.

“We could… verify that the energy transfer is through a sound pulse,” co-author Luis Melo from Lisbon University’s physics department told AFP by email.

This not only solves “an old, fundamental problem,” it also boosts understanding of other types of oscillators, he said.

Huygens is credited with making the first pendulum clock in 1656—the most accurate timepiece built until then, with an error margin of less than one minute a day. He later improved this to under 10 seconds.

In 1665, the scientist reported observing a strange phenomenon—two clocks hanging from the same structure would start swinging in unison, though in opposite directions.

The speed of a pendulum—a weight suspended from a string, rod or wire—is determined mainly by its length.

Attempts to explain the synchronisation have ranged from the impacts of tiny air currents to minute movements in the base on which the clocks are mounted.

Original report: Tic toc: Why pendulums swing in harmony.

  1. oldbrew says:

    A bit like this but without the linking string.

  2. gymnosperm says:

    And they synchronize in opposite directions?

  3. oldbrew says:

    This may or may not be the same phenomenon – you decide.

  4. mpcraig says:

    Definitely not related but it reminded me of this cool video: https://www.youtube.com/watch?v=yVkdfJ9PkRQ

    [reply] yes, worth a look

  5. Paul Vaughan says:

    Some of you may recall wuwt getting stats advice from a guy named Briggs who used to always say data should “never” be smoothed …but note how he later relented, acknowledging William Hsieh’s stern counsel about overfitting noise where exogenous volatility structure guides otherwise (i.e. away from untenable assumptions).

    OB’s article on synchronization is timely:

    A paper to which I’ve pointed before:

    Epochs of phase coherence between El Nino/Southern Oscillation and Indian monsoon

    Their section “3. Phase Reconstruction” concisely outlines some beautiful methods that will help the community learn that Tsonis & colleagues (2007 & 2011) overlooked key synchronizing factors outlined by Dickey & Keppenne (NASA JPL 1997 Figures 3a & 3b). (bidecadal heat engine asymmetry on the ENSO spatial mode)

    Maraun & Kurths’ (2005) methods can be refined, for example to further clarify the adjustment synchronization (NOAA ERSSTv4 “bias” (actually natural)) I’ve illustrated here:


  6. oldbrew says:

    Paul Vaughan says: July 27, 2015 at 7:16 pm

    Yes…’ it also boosts understanding of other types of oscillators, he said’.

  7. Boyfromtottenham says:

    As a clock collector, and past owner of a pendulum clock, I’m struggling to understand what the “sound pulse” is that is thought to cause the synchronicity. Pendulums per se do not make any noise, it’s the escapement that regularly releases the driving energy that causes the “tick”. Maybe these folk haven’t actually seen and heard a pendulum clock working!

  8. ivan says:

    Following on from Boyfromtottenham. Did they try enclosing one of the clocks in a sound proof box to test their assumption that it was sound that did the synchronising, or did they follow the climate science agenda and assume the results were because of the model?

    I can think of several other things that placing one clock in a sound proof box would also eliminate, moving air and possible vibrations being transmitted through the common mounting surface.

  9. oldbrew says:

    Found a link to the science paper:

    ‘The exact mechanism of how the pulse energy propagates through the clock hardware down to the pendulum is hard to assess in detail and depends on the individual clock.’

  10. Bob Greene says:

    Reblogged this on JunkScience.com and commented:
    Maybe I need some more pendulum clocks.

  11. jdmcl says:

    “A pair of Portuguese scientists hypothesised that these pulses might move from clock to clock, perturbing the swing of the pendulums and eventually causing them to synchronise. .. They developed a complex mathematical model before conducting experiments with a pair of clocks attached to a rail fixed to a wall.”

    This is as bad as climate modelling – they decided what they were looking for and then with that preferred outcome in mind, constructed a model and after that failed to properly test the negative situation (i.e. with clocks where the sound could NOT travel from one clock to the other).

  12. AlecM says:

    Pretty obvious, proved by finding how far they need to be apart for the effect to slow then vanish.

  13. Russ Wood says:

    And I believe that the US TV series Mythbusters tried this some years ago. They put a load of small clocks on a freely moving table, and watched as they, over time, synchronised their tick.

  14. oldbrew says:

    Russ W: the moving platform was the key. When they prevented the platform from moving in mid-experiment, the metronomes went out of sync right away.

  15. tchannon says:

    I’ve held off commenting on this article.

    Your last comment oldbrew correctly states a simple fact: coupled resonant devices are coupled.

    Talk of on a wall as though this is somehow of importance tends to figure without pointing out the wall moves. PR, news outlets which includes most journals don’t do no news, also obvious, most do fake, in effect mislead as though they know not what they write.

    Would it work with a Harrison chronometer? My point is that decoupling the timing mechanism from it’s frame of reference is necessary.

    The article claims there are things to be learnt about oscillators, a threadbare claim. There is a class of pseudo oscillator where knowledge is incomplete, chaotic systems.

    There are a variety of ways energy transfer can occur, air vibration is one so the only point of the work is some kind of solution for that case. It isn’t complete.

    Behind your post oldbrew is I suspect an interest in influence of vibrating systems including at the planetary level. I agree. So far as I know all of those are to a degree chaotic, a very complex and interesting situation.

    Chaotic resonance is highly sensitive to perturbation some of the time, when tipping onto a particular path is finely balanced.

    Works relatively recently showed that chaotic oscillators can be controlled, tripped to regular. There are other related ways as well, a field where I have experience except that explaining would take a lot of time and space, a basis is needed.

    Earth systems too are I suspect chaotic, modal. I’ve concluded modal is where extraterrestrial influence comes in. Can’t prove it.

    There are even chemical oscillators, I expect biological too, not expect, know, chaos and eg. heart triggers, pacemakers is well known. Are our thoughts chaotic at heart?


    In electronics here is an enormously cited author citing himself

    Your might like to try this word phrase, with US spelling, on Bing or Google, without quotes

    synchronized chaotic oscillator

  16. Paul Vaughan says:

    Point of clarification:
    The “sound” (vibration) is through the rail, not the air.

    Those following OB’s link will find that the study is not paywalled.

    With regards to the geophysical context:

    Synchronized chaos?
    Yes! Exactly.


    This is the intersection of aggregate cyclic volatility of:
    A. heliocentric (J+N) equator-pole semi-annual heat engine.
    B. barycentric (J-N) interhemispheric annual heat engine.

    Keep in mind that the Findlater Jet is an interhemispheric kingpin.

    “Figure 5. Phase difference of ENSO and Monsoon (black). Grey shading marks intervals of jointly well defined phases. 1886–1908 and 1964–1980 (I): plateaus indicate phase coherence. 1908–1921, 1935–1943 and 1981–1991 (II): Monsoon oscillates with twice the phase velocity of ENSO. During these intervals, both systems exhibit distinct oscillations (NINO3 time series, upper graph). 1921–1935 and 1943–1963: phases are badly defined, both processes exhibit irregular oscillations of low variance (upper graph). Lower graph shows volcanic radiative forcing index (VRF).”Maraun & Kurths (2005)

    Sometime maybe this week I’ll see if I can dig out the link for Hsieh’s seminar where he sagely counsels against allowing overly sensitive Hilbert transforms to get kicked off-track by culturally-pervasive false methodological assumptions. I’ve been really disappointed that the climate discussion community can’t take methodological discussion to the philosophical level. The discussion (for example at wuwt & ce) is polluted by “contributors” who want to keep the discussion mired in technicalities based on false assumptions that fail diagnostics. Eventually all of that distortion will get corrected by higher powers — and by that I don’t mean more aggressive thought police, but rather more lucid cognitive powers.

    The cure for darkness is enlightenment …and every day it’s becoming more & more clear that it will be administered to the climatically darkened west by the naturally enlightened east.

    I advise everyone to carefully study Maraun & Kurths’s (2005) section “3. Phase Reconstruction”. I’ll be illustrating refined application of those methods in the coming months.

  17. hunter says:

    They are coupled by that which connects them. Isolated properly and they will very likely not synch.

  18. oldbrew says:

    Synchronisation on longer timescales: planetary example here (Neptune:Pluto).

    ‘Second, the longitudes of ascending nodes of the two bodies—the points where they cross the ecliptic—are in near-resonance with the above libration. When the two longitudes are the same—that is, when one could draw a straight line through both nodes and the Sun—Pluto’s perihelion lies exactly at 90°, and hence it comes closest to the Sun when it is highest above Neptune’s orbit. This is known as the 1:1 superresonance. All the Jovian planets, particularly Jupiter, play a role in the creation of the superresonance.’

  19. Paul Vaughan says:

    hunter (July 28, 2015 at 6:44 pm) suggested:
    “They are coupled by that which connects them. Isolated properly and they will very likely not synch.”

    You don’t have to speculate. Just look at I & II on the graph. (it’s simple)
    Now with 20/20 hindsight: ENSO-ISM coupling is NOT independent of the heat engines.
    (what the graph shows is hardly surprising if you think about it)

    [reply] should ‘ISM’ read ‘IMS’ for Indian Monsoon Season?

  20. Paul Vaughan says:

    ISM = Indian Summer Monsoon

  21. oldbrew says:

    ‘South Asia monsoon: Analysing fresh water could be key to forecast’

  22. Zeke says:

    I don’t have a problem with pendulums attached to the same wall. What keeps me up at night are falling slinkies in slow motion.