Listen

The Ohm Store Research

Why Singing Bowls Pulse: Revealing The Fundamental Doublet

For the last 50 years, we’ve (yes us too) as an industry been referring to the fundamental notes of bronze singing bowls incorrectly.

“What note is this bowl?” is a useful question, but it’s based on a fundamental misunderstanding of these instruments.

Bronze bowls usually carry a fundamental double. Two close tones, rather than one, defines the “note” of a handmade singing bowl.

Thousands of questions and comments over the years have begun with – “What note is it?”

It's the same question we ask of a piano key, a guitar string or a crystal bowl.

And it’s totally fair! Because a bronze bowl does have a prominent note, and we use that musicality in everything we do.

But it's only half the story. A handmade bronze bowl doesn't sing one fundamental note. It sings two, a fundamental doublet: two nearly identical tones that weave together into the slow, living pulse you feel in your body.

The Big Takeaways

  1. A note name is useful — and incomplete. Bronze bowls have two close fundamentals, not one.
  2. Those two tones beat. You hear one musical pitch whose loudness rises and falls — the pulse.
  3. In our archive, 95.6% of measurable takes showed a clear fundamental doublet (median ~1.9 Hz apart → ~2 pulses/sec).
  4. We recorded and analyzed 1,852 bowl recordings for this study, including 1,702 single-bowl takes.
  5. Of those singles, 1,494 showed a clear fundamental doublet. Zero showed clear evidence of a fundamental without one.
  6. Of the 1,562 takes long and clean enough to judge, 95.6% were clear doublets.
  7. Typical partners sit 1.89 Hz apart (median; middle half 1.15–2.73 Hz) — about 14.7 cents, less than a quarter of a semitone.
  8. About 86% of clear doublets split by less than 25 cents, and nearly all (~99%) by less than half a semitone — so you hear one note that moves, not two.
  9. That gap writes a pulse of about twice a second (median beat 1.82 Hz; median period 0.55 s). Most clear bowls fell between about one and three pulses per second.
  10. In about seven of ten clear cases (68.8%), a higher partial near 3× the fundamental was a doublet too — a quieter, quicker shimmer above the slow body breath.

What note is this bowl?

Ask most sellers what a bowl “is,” and you’ll hear a note name: a G, an E, a D#.

That answer is useful for musicality and for matching bowls!

A bronze singing bowl’s fundamental is usually two peaks, extremely close together.

The ear still “hears” a pitch, but it feels like it’s one pitch wavering up and down. This is actually tone tones, modulating between them!

Crystal bowls are often much closer to a single, clean fundamental (with their own overtone story). Hand-hammered bronze almost never is. The illustration below puts the difference in one glance.

Traditional pitched instrument versus the doublet of a bronze bowl — one clean peak compared with Tone A and Tone B

Hear it for Yourself Now

One of our new Reference Grade Bowl, a “G3.” First the real strike, then each fundamental tone alone, then both together. Alone, each tone is steady. Together, you hear the true modulations between the two notes.

Fundamental doublet: 195.9 Hz + 197.1 Hz → pulse 1.22×/sec

These are the bowl’s two lowest tones only — the full bowl rings with many higher tones too.

🔊 Tap a button to listen (turn your sound on)

Now playing: —

Feel the gap

Drag how far apart the two fundamental tones are (0–4 Hz). Watch the loudness swell, then listen. This is the fundamental doublet only — not the whole bowl.

Why two tones?

The bowl vibrates in two nearly identical patterns, turned slightly apart. Each has its own pitch — close enough to fuse into one note, far enough to pulse. Those two pitches are the fundamental doublet (the lowest pair), not every tone the bowl can make.

Five short clips from different Reference bowls. Each label is the measured fundamental-doublet pulse rate (from the two lowest tones only).

Press play: you’ll see and hear each bowl’s two lowest tones beating into a pulse.

Three Reference bowls. Each frame shows the fundamental doublet only (the two lowest tones that make the pulse — not every tone in the bowl), with a big pulsing ring synced to the sound.

What do you hear when a singing bowl “breathes”?

Invite a bronze bowl to sing, and you’ll hear a swelling and rising and building and decaying series of tones and overtones.

It feels as if the bowl is breathing.

We can call it the “wah-wah”, pulse, throb, shimmer, “life in the tone.” The bowl feels alive, and this is why…

Two close tones → one fused pitch + a loudness pulse at their difference.

The cause, in the cases we study here, is ordinary and precise. The bowl’s lowest ringing mode is not one frequency but two, extremely close together.

Those two tones sound at once.

Because they differ by only a hertz or two, it’s not obvious as you listen that it’s two distinct tones or frequencies. It feels like one pitch whose loudness rises and falls. .

We call that pair a fundamental doublet. Following we share – how large the split usually is, how fast the resulting pulse is, and how often a similar split appears in a higher partial.

What is a singing bowl doublet?

A doublet is a pair of nearly identical partner tones that belong to the same vibrational mode, but sit a tiny frequency gap apart.

Played together, they produce beats at that gap rate. We hear one note that pulses. The “one” peak resolves into two.

Figure 1: Spectrum of a bronze singing bowl showing two close fundamental peaks about 1.6 Hz apart, and the amplitude envelope pulsing about every 0.64 seconds
Figure 1. A real bronze bowl’s fundamental doublet (the two lowest tones only — not the whole bowl). (a) High-resolution spectrum near the lowest mode, showing two peaks. (b) Amplitude envelope of that band after the strike, showing the audible pulse (“wah-wah”).

Key findings

  • 95.6% of measurable single-bowl takes showed a clear fundamental doublet (1,494 of 1,562 testable recordings).
  • Typical partner split: median 1.89 Hz (IQR 1.15–2.73 Hz), or about 14.7 cents — less than a quarter semitone.
  • That gap makes an audible pulse of about two beats per second (median beat 1.82 Hz; median period 0.55 s).
  • In about seven of ten clear cases (68.8%), a higher partial near 3× the fundamental was also a doublet.
  • Zero clear “no doublet” bowls among 1,702 single-bowl recordings.

A singing bowl’s “wah-wah,” pulse, or wobble is usually two nearly identical frequencies in the lowest mode beating against each other — a fundamental doublet. Physics of imperfect circular shells predicts this for every real hand-made bronze bowl.

One number to keep: median split 1.89 Hz → median pulse 1.82×/sec (about twice a second).

In plain terms: when a bronze singing bowl is struck, many listeners hear not a frozen tone but a slow, living pulse. Across thousands of bowls we’ve recorded and analyzed — more than 1,800 recordings, 1,702 single-bowl takes — we found clear fundamental doublets in 1,494 cases and no clear absences. Among takes long and clean enough to judge, 95.6% showed the doublet. The partners typically sit ~1.9 Hz apart and beat about twice a second. That is singing-bowl physics you can hear.

Go deeper — shell modes, degeneracy, and why asymmetry splits them

Imagine a perfectly round bronze bowl, every wall equally thick, every alloy grain the same. In that ideal object, the two (2,0) rim shapes would ring at exactly one frequency. Nature does not cast that object. Hand-hammered bowls are slightly out of round. The wall is a little thicker here than there. The alloy is not perfectly uniform. Hammer marks and the lip’s finish break the last bits of symmetry.

Those tiny asymmetries do something specific to a degenerate mode pair: they lift the degeneracy. One partner’s frequency rises a little; the other’s falls a little, or rises less. The angular shapes remain roughly a quarter-turn apart for the fundamental, but the frequencies no longer match. The result is a doublet, and the audible consequence is beating.

The same story is older than singing-bowl research. Work on ancient Chinese two-tone bells — deliberately non-circular — showed otherwise matched modes split into doublets whose members can differ by several percent. Oval cylinders and noncircular bells do the same. Standard treatises on musical instruments treat warble and split modes in bells as the everyday consequence of broken symmetry. Laboratory measurements of Tibetan bowls found that even high-quality handcrafted bowls are never perfectly axisymmetric: what looks like one resonance often resolves, under zoom, into two closely spaced peaks belonging to orthogonal modal families.

None of this requires a flaw in the usual sense. It requires only that the bowl be real. A split of one or two hertz on a 200 Hz fundamental is a fractional change of about one percent or less — far smaller than the intentional two-tone splits of Chinese bells, but large enough to write a pulse into the sound. Hand-making does not fight this physics. Hand-making is this physics.

Go deeper — how we measured beating (methods)

We recorded and analyzed thousands of bronze singing bowls. The working archive for this study contained more than 1,800 audio recordings. After setting aside multi-bowl and set tracks, which mix several instruments into one file and cannot answer a single-bowl question cleanly, 1,702 single-bowl recordings remained.

For each recording we located the principal strike, then examined several seconds of the ring that followed — typically up to about fourteen seconds when the file allowed. In that window we computed a high-resolution spectrum of the decay and searched the fundamental region for a close peak pair. Separately, we filtered a narrow band around the fundamental, took the amplitude envelope of that band, and measured any periodic loudness modulation (the beat). A recording was classed as a clear fundamental doublet when we resolved a credible peak pair, or when a clear periodic beat was present even if the spectrum had not fully separated the peaks. A recording was classed as no doublet only when the take was long enough that an ordinary split would have shown, yet neither a pair nor a beat appeared.

One limit matters for every claim that follows. The slowest beat a recording can reveal is about one cycle over the length of the analysis window. On a typical clean take of roughly thirteen to fourteen seconds, beats much slower than about 0.07 Hz (a pulse taking many seconds per cycle) are invisible. A bowl with an extremely slow split could therefore look, in a short file, like a single peak. We treat that limit openly in the results and the conclusion.

How often is the fundamental a doublet?

Of 1,702 single-bowl recordings, 1,494 showed a clear fundamental doublet.

Zero showed clear evidence of a fundamental without a doublet.

Among the 1,562 singles long and clean enough to test, 95.6% were clear doublets.

Among measurable takes: 95.6% clear doublets. No clear “single-tone” population.

Finding 1 — The two tones are typically about 1.9 Hz apart

Among the 1,494 clear doublets, the frequency gap Δf had a median of 1.89 Hz. The middle half fell between 1.15 and 2.73 Hz.

In musical cents, the median split was 14.7 cents (IQR 8.8–20.6) — less than a quarter semitone.

About 86% of clear doublets split by less than 25 cents, and nearly all (~99%) by less than half a semitone. That is why the experience is one note, not two.

Ear check: 220.0 Hz + 221.9 Hz (Δf = 1.9 Hz) is not “two A’s” — it is one A that pumps about twice a second. That is our median bowl.

A tuner looking at a coarse spectrum sees one peak. A player in the room hears a living tone. Both are right. The pulse was the doublet’s signature all along.

Go deeper — why the ear fuses the partners

The auditory system is excellent at hearing beats and roughness from close frequencies, and poor at labeling those close frequencies as separate pitch names when they lie well inside a fraction of a semitone and share the same attack.

The “one” peak stays one peak until frequency resolution is fine enough, or the window long enough, to pry it into two.

Finding 2 — That gap makes a pulse of about twice a second

Beats are not a metaphor. The beat rate equals the frequency difference: a 1.8 Hz split pulses at 1.8 Hz on a low bowl or a high bowl alike.

In our clear doublets the median beat rate was 1.82 Hz (median period 0.55 s) — roughly two pulses a second. The middle half beat between 1.09 and 2.68 Hz. Most bowls fell between about one and three pulses per second.

What those rates feel like:

  1. ~1×/sec — slow and spacious; you can nod with each swell.
  2. ~2×/sec — our median; a gentle throb you feel without counting.
  3. ~3×/sec — a quicker shimmer, still motion inside one note.
Go deeper — the beat-rate identity

If two steady tones of equal strength have frequencies f and f + Δf, their sum is a tone at the average frequency whose amplitude envelope repeats exactly |Δf| times per second. There is no extra constant and no dependence on absolute pitch.

Figure 2: Histogram of fundamental beat rates for 1,494 clear singing bowl doublets with median 1.82 Hz and most bowls between one and three pulses per second
Figure 2. Distribution of fundamental beat rates among 1,494 clear doublets. The dashed line marks the median (1.82 Hz). The shaded band marks about one to three pulses per second.
Go deeper — Finding 3: higher partials as doublets (~7 of 10)

Above the fundamental, bronze bowls usually carry a strong higher partial near three times the fundamental frequency — the next member of the same flexural family, often labeled (3,0). In 68.8% of our clear-fundamental cases, that higher region also showed doublet structure: a resolved peak pair, a clear envelope beat, or both. In round language, about seven of ten bowls.

Why should higher modes split too? For the same reason the fundamental does. Each (n,0) flexural mode of a nearly circular shell is born as a degenerate orthogonal pair. Any asymmetry that lifts the (2,0) degeneracy is available to lift the (3,0), (4,0), and higher pairs as well. A bowl that is slightly oval or unevenly thick is not “asymmetric only for the fundamental.” It is asymmetric for the shell.

Why, then, do we detect the upper doublet less often than the fundamental one? Not because physics switches off. Because measurement gets harder. The higher partial is typically much quieter than the body tone — often tens of decibels down — so its partner sits nearer the noise. It also usually decays faster, leaving a shorter clean window for fine frequency resolution. Our 68.8% figure is therefore a lower bound on how often upper doublets exist, and a fair estimate of how often they declare themselves in ordinary strike recordings.

Perceptually, the upper doublet is the “faster, quieter shimmer” many listeners notice above the slow body pulse.

Go deeper — Finding 4: published acoustics already predicted this

Our counts are new as a large listening-archive census. The underlying mechanism is not. Laboratory work on Tibetan bowls has already shown split (2,0) peaks and audible beating from asymmetry. The wider acoustics of bells and circular shells has treated mode doublets as the normal fate of broken axisymmetry for decades. What our archive adds is scale and plainness: across thousands of bowls we’ve recorded and analyzed, the predicted doublet is not occasional. It is the rule.

Those papers do not say that every commercial recording will resolve the split. They say that the vibrational physics of a non-ideal axisymmetric shell produces partner modes with slightly different frequencies. Hand-made bronze bowls are non-ideal axisymmetric shells. Therefore partner splits are expected. Our measurements find them wherever the recording is good enough to look.

Why has the singing bowl doublet been easy to overlook?

If the effect is common, why do so many descriptions stop at “wah-wah” without naming two tones? Several reasons stack:

  1. The split is small — typically ~15 cents. The ear fuses partners into one pitch and reports loudness motion.
  2. Ordinary displays hide the pair. Short FFT windows smear two peaks a hertz apart into one mound.
  3. Bowl talk often stops at the note name (“a G”) without asking whether that G is one line or two.
  4. Everyday words already work — pulse, beat, life — they just stop one step short of mechanism.

Our aim is not to replace “wah-wah” with jargon. It is to let “wah-wah” point at something exact: a fundamental doublet.

How to choose a singing bowl: listen for the living pulse

The breath in the tone is not a studio trick or a microphone artifact. It is the bowl’s lowest mode showing that the metal is a real, slightly irregular shell — which is what hand-made bronze is.

A slow, even pulse near one to three times a second is, in our archive, a common fingerprint of that fact.

How to listen (simple steps):

  1. Listen through the decay, not only the attack.
  2. Ask whether the fundamental feels alive.
  3. Notice whether a quieter, quicker shimmer rides above it.
  4. Prefer takes — in person or in recordings — long and quiet enough for the pulse to declare itself.

Because the median split is so fine, do not expect to “hear two notes.” Expect to hear one note that moves.

None of this ranks bowls by spirituality, and none of it says a bowl without an obvious pulse in a short recording is “dead.” A bad take can hide a real doublet. When the breath is there, a fundamental doublet is the most parsimonious explanation — and the one our measurements support at scale.

Bowls we have already tuned and selected for everyday listening live in our curated collection of tuned bowls. Bowls that pass the fuller acoustic bar we use for reference-grade listings — purity, pulse, sustain, and balance — will appear in Reference Grade Singing Bowls as that collection is filled.

Hear bowls chosen for a living fundamental pulse.

Browse Reference Grade Singing Bowls

Takeaways

  • A note name is useful; a bronze fundamental is usually a doublet.
  • 95.6% of measurable takes showed a clear fundamental doublet; we found no clear single-tone population.
  • Typical split ~1.9 Hz (~15 cents) → pulse ~twice a second.
  • You usually hear one moving note — not two named pitches.
  • Listen through the decay. Prefer a clear, even breath in the body tone.

A closing word

None of what we're describing here can measure whether a doublet has any impact on the therapeutic effects of a bowl.

We have played thousands of bowls, and all have their own purpose. It is the player who creates a cohesive experience with other instruments, their own mindset and intentions, and how they use the bowl. That is what ultimately determines its effectiveness.

Our Reference bowls are simply chosen to showcase a particular, special quality of bowls, not because they are "more therapeutic."

FAQ: singing bowl beat, pulse, and “wah-wah”

Why do singing bowls pulse or make a “wah-wah” sound?

Usually because the lowest ringing mode is a fundamental doublet: two nearly identical frequencies beating against each other. The ear hears one note whose loudness rises and falls at the difference rate (often about twice a second).

Is “what note is this bowl?” enough?

It is useful for musicality, but incomplete for bronze. The fundamental is usually two close tones, not one clean peak like a typical string. Crystal bowls are often closer to a single fundamental.

What is a singing bowl doublet?

A pair of partner tones belonging to the same vibrational mode (for the fundamental, the (2,0) flexural mode) whose frequencies have been slightly split by real-world asymmetry in the shell.

How fast is a typical singing bowl beat?

In our 1,494 clear doublets, the median beat rate was 1.82 Hz — roughly two pulses per second (median period 0.55 s). Most bowls fell between about one and three pulses per second.

How common is beating in Himalayan / Tibetan singing bowls?

Very common in our bronze archive. Among 1,562 measurable single-bowl takes, 95.6% showed a clear fundamental doublet. We found no clear population of bowls without one.

Can you hear two notes, or just one pulsing note?

Usually one pulsing note. The typical split is about 15 cents — less than a quarter of a semitone — so the ear fuses the partners into one pitch and reports the conflict as loudness motion.

What makes a good singing bowl sound?

Listen through the decay, not only the attack. A clear, even fundamental pulse (and sometimes a quieter, quicker shimmer above it) is a common fingerprint of a real hand-hammered shell. Prefer long, quiet takes so the pulse can declare itself.

How should I choose a singing bowl by ear?

Strike once, wait for the impact flash to fade, and listen for life in the body tone. Ask whether the fundamental feels alive. Compare several bowls side by side when you can. Our Reference Grade collection will gather bowls that passed our acoustic tests; our curated tuned bowls are already selected for everyday listening.

Is the pulse a recording artifact or “added” effect?

No. Beating follows from two close frequencies sounding together. Published acoustics of bells and Tibetan bowls predicted this for imperfect circular shells long before our census.

Do higher tones in a bowl pulse too?

Often. In 68.8% of our clear-fundamental cases, a higher partial near three times the fundamental also showed doublet structure — a faster, quieter shimmer above the slow body breath.

How does singing bowl physics relate to bells and wineglasses?

The same shell physics: broken symmetry lifts degenerate flexural modes into doublets. Classic work on Chinese two-tone bells, noncircular cylinders, wineglass acoustics, and Tibetan bowls all point at partner splits; our contribution is a large plain-language census of bronze bowls we’ve recorded.

Glossary

Fundamental doublet
Two close frequencies belonging to the bowl’s lowest (2,0) flexural mode; their difference produces the audible pulse.
Beat / pulse / “wah-wah”
Periodic loudness rise and fall at rate |Δf|, heard as breath or wobble inside one note.
Δf (delta-f)
Frequency gap between partner tones, in hertz. Median ~1.89 Hz in our clear doublets.
Cents
Hundredths of a semitone. Our median split ~14.7 cents — less than a quarter semitone.
(2,0) mode
Lowest flexural rim mode of a circular shell: four circumferential nodes, no nodal circle up the wall.
Degenerate modes
On a perfect round shell, two orthogonal partners locked at the same frequency; asymmetry splits them into a doublet.
Upper / higher doublet
Similar partner split on a higher partial (often near 3× the fundamental); heard as a quieter, quicker shimmer.

References

  1. Fletcher, N. H., & Rossing, T. D. (1998). The Physics of Musical Instruments (2nd ed.). Springer. https://link.springer.com/book/10.1007/978-0-387-21603-4
  2. French, A. P. (1983). In vino veritas: A study of wineglass acoustics. American Journal of Physics, 51(8), 688–694. https://doi.org/10.1119/1.13147
  3. Inácio, O., Henrique, L., & Antunes, J. (2006). The dynamics of Tibetan singing bowls. Acta Acustica united with Acustica, 92, 637–653. Related open write-up: https://www.sea-acustica.es/archivo/publicaciones/revista_VOL35-12_05_01.pdf
  4. Rossing, T. D., Hampton, D. S., Richardson, B. E., Sathoff, H. J., & Lehr, A. (1988). Vibrational modes of Chinese two-tone bells. Journal of the Acoustical Society of America, 83(1), 369–373. https://doi.org/10.1121/1.396250
  5. Rossing, T. D., & Hansen, U. J. (1993). Modal analysis of noncircular cylinders and bells. Journal of the Acoustical Society of America, 94, 1806. https://pubs.aip.org/asa/jasa/article/94/3_Supplement/1806/733585/Modal-analysis-of-noncircular-cylinders-and-bells
  6. Terwagne, D., & Bush, J. W. M. (2011). Tibetan singing bowls. arXiv:1106.6348. https://arxiv.org/abs/1106.6348; PDF; author PDF https://thales.mit.edu/bush/wp-content/uploads/2012/04/TibetanBowls.pdf

© The Ohm Store Research. Measurement study of bronze singing-bowl acoustics. No health or healing claims.