Playing harmonics on a stringed instrument is an acknowledged technique which involves placing a finger at a certain point on the string to damp a node that would otherwise sound.

Is there a counterpart technique to playing a wineglass with a wet finger? Can you damp a node by touching it at particular point or points on the glass?

I've tried but I've only been able to get a single note out of a wineglass, and I wondered whether I'm not doing it right or whether it's actually possible at all.

I'm imagining a choir piece with ethereal "bowed" notes from wineglasses, and I wondered how many sounds I could reasonably achieve.

  • I think there are some fundamental physics problems. A string can be touched at a "point" because the string is a 2D line, and we find points on a line. The wineglass is a 3D cylinder! The vibration passes through all points: youtu.be/5VZcY1B8iPM Commented Jun 17 at 12:30
  • 1
    @AndyBonner But it is possible to promote harmonics on a Chladni plate, which is essentially a 2D resonator like the glass (I would argue that a string is a 1D object, not 2). Hence there is probably nothing preventing it. The main problem being that, as the geometry is complex, the possible modes are probably fairly hard to determine.
    – Tom
    Commented Jun 17 at 12:36
  • @Tom Oh, right, math hiccup. Meanwhile Brian, there's a reason wineglasses are typically played in sets! You might also explore instruments with similar sounds like bowed psaltery. (The latter loses the "circular breathing" ability for infinite sustain, though.) Commented Jun 17 at 12:57
  • 1
    I foresee two problems: 1) I play piano and trumpet, neither of which really does harmonics, and 2) I don't think most wine glasses would support my weight.
    – Aaron
    Commented Jun 17 at 18:28
  • There must be nodes on a glass. The trick is to find them.
    – user207421
    Commented Jun 18 at 7:40

3 Answers 3


Glasses suitable for singing are designed to have a rather pure tone. Because the bending modes are comparatively complex (like those of free reeds are), the overtones tend to be rather disharmonic, namely not in a useful relation to the fundamental. That means that glasses with a nice voice tend to have quite dampened (and disharmonic) overtones.

That in turn means that you'll have a hard time finding a way to dampen the fundamental to a degree where you can excite an overtone enough to become audible over it. And it will typically not be in a recognizable harmonic relation to the fundamental.

The normal way to "tune" glasses of similar size is to fill them to various degrees with water (to get lower, you can use salt water which has a higher density without significantly increased viscosity).

  • 2
    I prefer to fill mine with spirits, thus having an extra reward at the end of the performance...
    – Tim
    Commented Jun 17 at 13:32
  • @Tim Why wait until the end :P?
    – Tom
    Commented Jun 17 at 18:21
  • 1
    @Tom - simple, it's not nice playing on out-of-tune instruments!
    – Tim
    Commented Jun 18 at 6:31
  • Another point is that a purpose-built glass harmonica that was ground to establish a permanent tuning might be ground in a way that creates useful harmonics, but filling a glass with water would change the relationship between harmonics and the fundamental.
    – supercat
    Commented Jun 19 at 14:14

A wineglass is, from the physics point of view, essentially a 2D object: its thickness is very small compared to its other dimensions. It follows that it should display some partials, aka modes, which are not necessarily in an harmonic fashion (like a bell for instance), but it should display some.

The situation is actually closer to a Chladni plate, with the important difference that a Chladni plate is usually as wide as it is large, making the determination of its modes easier. On a Chladni plate, it is fairly easy to promote one mode over another one, by bowing at a specific position and preventing the plate vibration in some places (nodes) with fingers. This is for instance done in this video.

Now, a glass does exhibit different modes, as it can be clearly seen in the video linked by Andy Bonner, but these modes are essentially on the circumference of the glass. Like for a Chladni plate, the potential modes you can trigger are dependent on the point of excitation, which is unfortunately moving when you are making a glass sing, hence the position of the ideal nodes are also moving.

If you have a bow, might be worth trying actually bowing the glass, and try to find some other modes by placing your fingers on the circumference.

  • 1
    I wonder if touching a glass harmonica firmly at one point but loosely at another would be a practical way of producing harmonics? Rotating one's finger around a stationary bowl would result in nodes and antinodes that would rotate at the same rate, but rotating two fingers around the bowl while maintaining a consistent phase angle would be difficult. Two stationary fingers on a spinning bowl would seem much easier.
    – supercat
    Commented Jun 18 at 17:46
  • @supercat Very good point! If you give me access to one I promise to try ;)!
    – Tom
    Commented Jun 19 at 11:39

There exists an instrument called glass harmonica, which uses one glass or - more correctly - bowl per tone. So I doubt that even if a different method existed for creating more than one tone per glass, it is reliable or quick enough to be of use under concert conditions.


Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.