Coupling between two singing wineglasses

Tal Arane, Ana K. R. Musalem, and Moti Fridman

Dept. of Physics of Complex Systems,
Weizmann Institute of Science, Rehovot 76100,
Israel
(Dated: January 8, 2009)
Coupling between two singing wineglasses was obtained and investigated. Rubbing the rim of
one wineglass produce a tone and due to the coupling induces oscillations on the other wineglasses.
The needed coupling strength between the wineglasses to induce oscillations as a function of the
detuning was investigated.

PACS numbers: 43.20+g, 62.30+d

arXiv:0901.0656v2 [physics.pop-ph] 8 Jan 2009

A wineglass can be made to sing by gently rubbing

its rim with a moist finger. The friction between the fin- scope
ger and the rim of the wineglass causes the wineglass to detector
oscillate, to produce a loud, pure tone [1]. The frequency
of the oscillations depends on the volume of liquid in-
side the wineglass, whereby adding liquid increases the
fluid pressure that retards the wineglass vibration so as glass
glass #1#1 glass #2
glass #2

to lower the frequency [2]-[3]. water

Here we investigate how two singing wineglasses can detector container
be coupled to each other. Specifically, we show that mirror
when the natural oscillating frequency of each wineglass
is comparable to each other, it is relatively easy to in- mirror
BS
duce oscillations of the same frequency and phase from
laser
one wineglass to the other. However, when their natural
oscillation frequencies differ, it is more difficult to induce
the oscillations, if at all. Such coupling plays an impor- FIG. 1: Experimental arrangement for investigating the cou-
pling between two wineglasses.
tant role in many ensembles such as coupled bubbles [4],
coupled lasers [5], etc.
The experimental arrangement is schematically shown
amplitudes and phases for the rubbed wineglass (driv-
in Fig. 1. Two wineglasses were submerged in a water
ing wineglass) and the unrubbed wineglass (driven wine-
container, and the oscillating frequency of each wineglass
glass) is shown in Fig 2. As evident, the frequencies and
was obtained by illuminating them with laser beams de-
phase of both the driving and induced oscillations are
rived from a HeNe laser and detecting the reflected light.
essentially identical. Clearly indicating that frequency
In this arrangement the ease of introducing oscillations
locking and phase locking accrue.
from one wineglass to the other, i.e, the coupling strength
between the wineglasses depends on the distance between 3
x 10
them, on the water level in the container and on the dif- 1
ference between the natural oscillation frequencies of the
Amplitude [V]

0.5
wineglasses (detuning).
We began our experiments by measuring the natural 0

frequency of each wineglass as a function of volume of 0.5

water inside them. We found, as expected, that the nat- 1

ural oscillating frequency decreases from 700HZ to 450HZ
as the volume of water inside them increases. Then, we 1.5
0.225 0.23 0.235 0.24 0.245 0.25 0.255 0.26 0.265
investigated the behavior and content of the induced os- Time [sec]
cillations as compared to the detuning oscillations, when
the water level inside each was the same as the other. FIG. 2: Oscillation frequencies and amplitudes for two cou-
This was done by rubbing the rim of only one wineglass pled wineglasses. Dashed curves denote the driving wineglass
to produce a natural oscillation frequency with a certain and solid curves the driven. The level of the water in the con-
amplitude and measuring the amplitude of the oscillation tainer was 13cm. The distance between the wineglasses was
frequency that is induced in the other wineglass. The 5mm.
coupling strength between the wineglasses corresponds
to the ratio of amplitudes of the oscillation frequencies. In order to determine the influence of distance between
A representative example of the oscillation frequencies, wineglasses and detuning, we repeated the measurement
 2

above at different distances and different detuning con- FIG. 4: Ratio of amplitudes of the two wineglasses oscilla-
ditions. The results are presented in Figs 3- 5. Figure 3 tions as a function of detuning between them. The distance
shows the coupling strength as a function of distance be- between the wineglasses was 25mm, and the water level in
tween the wineglasses, for different levels of water in the the container was 12cm.
container. As evident, the coupling strength decreases
significantly as the distance between the wineglasses in-
creases. On the other hand, the influence of water levels
of water in the container is not significant.
0.9
Finally, Fig. 5 shows the critical detuning as a function
0.8
of coupling strength, for different levels of water in the
container. As evident, when the coupiling strength which
Coupling Strength

0.7 height 11 cm
height 12 cm
induce oscillations in the driven wineglass increases, the
0.6
height 13 cm critical detuning increases as well. Also, when the level of
0.5
water in the contained increases the slops of the curves in-
0.4
crease, probably due to changes in the damping strength
0.3
which depend on the water level in the container
0.2

0.1

0
0 5 10 15 20 25 30 35

Distance between wineglasses [mm]

40

FIG. 3: Coupling strength between two wineglasses as a func-

tion of the distance between them. 30
Critical detuning [HZ]

Figure 4 shows the coupling as a function of the detun- 20

ing between them, for a certain level of water in the con-
tainer (13cm) and a certain distance between the wine- 10
glasses (25mm). The detuning was achieved by simply Container water level
varying the volume of water inside one of the wineglasses. 11 cm
0 12 cm
As evident, the induced frequency amplitude decreases as
13 cm
the detuning is increased. There is a point of critical de-
10
tuning which indicates that induced oscillations can no 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9
longer be achieved in the driven wineglasses at a certain Coupling strength

coupling strength.
FIG. 5: Critical detuning as a function of coupling strength
0.2 for different water levels in the container.
0.18
Ratio of amplitudes

0.16

0.14

0.12

0.1
Critical detuning
0.08
To conclude, the critical detuning of two wineglasses
0.06

0.04
is directly related to the coupling strength between them
0.02
and vice versa. For higher coupling strength the critical
0 detuning is higher. For large detuning, it is necessary
0 1 2 3 4 5 6 7 8 9
to increase the coupling strength in order to ensure that
Detuning [Hz] induced oscillations will accure in the driven wineglass.

[1] T. D. Rossing, Wine glasses, bell modes, and Lord Anisotropy in the sound field generated by a bubble chain,
Rayleigh, Phys. Teacher 28, 582 (1990). Journal of Sound and Vibration 278 807 (2004).
[2] Y. Y. Chen, Why does water change the pitch of a singing [5] M. Fridman, V. Eckhouse, N. Davidson, and A. A.
wineglass the way it does?, Am. J. Phys. 73, 1045 (2005). Friesem, Efficient coherent addition of fiber lasers in free
[3] K. W. Chen, C. K. Wang, C. L. Lu and Y. Y. Chen, space, Opt. Lett. 32, 790 (2007).
Variations on a theme by a singing wineglass, Europhys.
Lett. 70 3 334 (2005).
[4] R. Manasseh, A. Nikolovska, A. Ooi, and S. Yoshida,