← Back to Learn
Soundb Learn · Studio Setup

What a Boundary Does to a Sound Wave

Reflection, absorption, diffraction and refraction — and the one piece of arithmetic that turns every room problem from a mystery into something you can predict with a tape measure.

Topic
Studio Setup
Level
Beginner
Format
Lesson
Time
13 min

Take a pair of monitors you know intimately, move them to another room, and they will sound like different speakers. Nothing about them has changed. What changed is everything the sound does after it leaves them.

This lesson is about that — what actually happens when a sound wave meets a boundary. It is the foundation for everything else in this specialization, and most of it comes down to a single piece of arithmetic that makes room problems predictable rather than mysterious.

The arithmetic that runs the whole subject

Start here, because nothing else in acoustics makes sense without it.

Sound travels at a speed set by the medium and, in air, by its temperature. The working approximation is:

v = 331 + (0.6 × T) — metres per second, with T in degrees Celsius.

At 20 °C that gives 343 m/s, which is the figure everyone quotes. In an Indian summer at 35 °C it is about 352 m/s — roughly 3% faster, which shifts every wavelength and every room-mode frequency by the same 3%. Worth knowing when a measurement taken in the morning does not quite match one taken in the afternoon.

From the speed you get the wavelength, which is the number that actually matters:

wavelength = speed ÷ frequency

Work a few out and keep them, because they explain more about rooms than any amount of listening will:

  • 20 Hz — 17 metres. Longer than most buildings.
  • 50 Hz — 6.9 m.
  • 100 Hz — 3.4 m. About the height of a room.
  • 250 Hz — 1.4 m.
  • 500 Hz — 69 cm.
  • 1 kHz — 34 cm. About the width of a laptop.
  • 5 kHz — 6.9 cm.
  • 10 kHz — 3.4 cm. About the width of two fingers.

Notice the range. Across the audible band, wavelength varies by a factor of a thousand — from longer than your house to shorter than your thumb. Almost every counterintuitive thing a room does follows from that one fact, because a physical object of a given size is enormous compared with a treble wavelength and invisible to a bass one.

Reflection

A wave travelling away from its source in an enclosed space eventually meets a surface, and what returns to the room is a reflection.

Off a flat, even surface it behaves like light off a mirror: the wave leaves at exactly the angle it arrived, measured from the surface. That is the angle of incidence rule, and it is what lets you predict where a reflection will go with nothing more than a straight edge and a drawing of your room.

How much comes back depends on two things. The material — the harder and less porous, the more is reflected and the less absorbed. And the frequency — higher frequencies are absorbed more readily than lower ones, which is why every reflection in a real room comes back duller than it went out, and why a reverb tail darkens as it decays.

The size rule, which is the important part

Here is the piece that most treatment advice leaves out, and it is the difference between spending money well and spending it badly.

A surface only reflects a sound effectively if the surface is comparable to or larger than the wavelength. If the wavelength is much longer than the object, the wave simply passes around it and carries on as though it were not there. The reflection is negligible.

Now put numbers on it. Take a standard 60 cm square acoustic panel:

  • At 4 kHz the wavelength is 8.6 cm. The panel is seven wavelengths across — enormous by comparison. It works extremely well.
  • At 1 kHz the wavelength is 34 cm. The panel is not quite two wavelengths across. It still works, less strongly.
  • At 100 Hz the wavelength is 3.4 metres. The panel is a fifth of a wavelength — effectively invisible. The wave goes round it.

That one calculation explains the single most common failure in home studios: a room covered in thin foam panels that has had all its life and air removed while the bass problems remain exactly as they were. The panels were never going to touch the bass. They are the wrong size, and no amount of them will fix it — which is why low-frequency absorbers are physically large objects and always will be.

Curved surfaces

Two shapes behave in ways worth recognising on sight.

A convex surface — bulging outward — scatters an incoming wave widely, because each part of it faces a slightly different direction. This is useful, and it is the basis of one whole family of acoustic treatment.

A concave surface — curving inward — does the opposite and focuses sound toward a point, exactly as a satellite dish does. That produces a hot spot where the sound is unnaturally loud and coloured, and a surrounding region where it is unnaturally weak. Domed ceilings, curved windows, arched alcoves and the inside of a bay window all do this. If a room has one, find the focal point and make sure it is not where anyone sits or where a microphone goes.

Absorption

Absorption is the reverse of reflection: energy that enters a material and does not come back, having been converted into a very small amount of heat.

Softer, more porous materials absorb more. And absorption is strongly frequency-dependent in the same way reflection is — high frequencies are absorbed readily by almost anything soft, while low frequencies need something substantial. This is the same size rule in a different guise, and it is why the two most common outcomes of amateur treatment are a room that is dull but still boomy, or a room that is dull and still boomy.

The mechanisms behind each family of absorber, and how to read an absorption coefficient table, are their own lesson later in this specialization.

Finding your reflection points with a mirror

The angle-of-incidence rule is not only theory — it gives you a two-minute technique that is more accurate than any amount of listening.

Sit in your listening position. Have someone slide a small mirror flat along the side wall while you watch it. Wherever you can see a speaker's drivers reflected in the mirror is a point where sound from that speaker reaches your ears after exactly one bounce. Mark it. Repeat for the other wall, the ceiling, and the desk surface in front of you.

Those marked spots are the first reflection points, and they are where absorption does the most good per panel. Everything works because light and sound obey the same reflection geometry — the mirror is doing the trigonometry for you.

Two things worth knowing while you do it. The desk is usually the worst offender and the one most people forget, because the reflection path is short and the surface is large and hard. And the ceiling point matters more in a small room than the side walls do, because ceilings are typically closer than walls and almost never treated.

Transmission: the third thing a boundary does

Reflection and absorption are two of three possibilities. The energy arriving at a surface has to go somewhere, and the three destinations always add up to what arrived:

  • Reflected back into the room.
  • Absorbed within the material and turned into a small amount of heat.
  • Transmitted through the boundary and out the other side.

That third one is the whole subject of isolation, and separating it from the other two prevents the most expensive mistake in home studios. A material can be excellent at absorbing — taking energy out of the room — and useless at transmission loss, because absorbing 80% of the energy at a surface still lets a great deal through and, more importantly, absorption works on the air while transmission is fought with mass.

Practically: foam and mineral wool change how a room sounds. They do essentially nothing about what the neighbours hear. Those are different problems with different physics, and the treatment course deals with the second one properly.

Diffraction

Sound does not only reflect and absorb. It bends around obstacles, and once again the amount depends on wavelength against object size.

A long wavelength meeting a small obstacle wraps around it almost completely and continues as though nothing were there. A short wavelength meeting a large obstacle is blocked, leaving an acoustic shadow behind it.

You have heard the consequence hundreds of times. Standing outside a venue, you hear the bass and the kick clearly and almost nothing of the vocal — the low frequencies are diffracting around and through the structure while the highs are being stopped by it. The same effect is why you can hear someone talking around a corner but cannot make out the words: the vowels carry, the consonants do not.

In a studio it explains why a low partition or a gobo screens the top end of a source effectively and does very little to its low end, and why isolating bass genuinely requires mass and structure rather than a barrier.

Refraction

The fourth behaviour, and the one people meet without recognising it.

Refraction is a change of direction caused by a change in the speed of propagation. Since the speed of sound depends on temperature, any temperature gradient in the air bends the path of a sound wave — it curves toward the cooler, slower region.

Outdoors during the day, the ground is heated by the sun and the air near it is warmer than the air above. Sound travelling near the ground is bent upward, away from a distant listener, and it seems to fade quickly.

At night the ground cools faster than the air, producing a temperature inversion, and the gradient reverses. Now sound is bent downward and returned to the ground, so it carries far further. That is the real reason a distant train, a highway or a wedding two streets away is so much more audible at night — not simply that the background is quieter.

Wind does the same thing by a different route, because wind speed increases with height: sound travelling downwind is bent toward the ground and upwind is bent away. It is why a sound system aimed downwind sounds fine and the same system into the wind seems to lose its top end.

Why treble is directional and bass is not

The size rule applies to sources as well as to surfaces, and it explains something you have certainly noticed.

A radiating surface spreads sound widely when it is small compared with the wavelength, and beams it forward in a narrowing cone when it is large. So the same driver is broadly directional at low frequencies and increasingly focused as frequency rises.

Put numbers on a typical two-way monitor:

  • A 25 mm tweeter at 10 kHz: the wavelength is 3.4 cm, so the dome is comparable to a wavelength and radiates over a reasonably wide angle. Push to 16 kHz and the wavelength drops to 2.1 cm, the dome starts to look large by comparison, and the output narrows.
  • A 20 cm woofer at 200 Hz: the wavelength is 1.7 m, eight times the driver, so it radiates almost spherically. At 2 kHz the wavelength is 17 cm — now the cone is larger than a wavelength and it is beaming hard, which is exactly why the crossover happens below that.

Three consequences follow. Your monitors have a much narrower usable listening window at the top of their range than at the bottom, which is why the on-axis position matters and why a step to one side dulls the treble but not the bass. Bass radiates in all directions, so it reaches every boundary in the room and excites everything — you cannot aim it. And a subwoofer's position is far less critical for imaging than a monitor's for exactly the same reason, though it is far more critical for modal behaviour.

Where this leaves you

Four behaviours, all sorting by wavelength:

  • Reflection — bounces off surfaces larger than the wavelength.
  • Absorption — takes energy out, easily at high frequencies and only with real mass at low ones.
  • Diffraction — bends around objects smaller than the wavelength.
  • Refraction — bends when the speed of propagation changes across the path.

What a room does to your monitors is the sum of all four happening thousands of times a second, and it is the largest single variable in your monitoring chain — larger than the speakers, larger than the converters, and considerably larger than the cables.

The good news is that none of it is mysterious. Every one of these behaviours is governed by wavelength, and wavelength is a division you can do on a phone. Once you know your room's dimensions and the speed of sound, you can predict where the trouble will be before you have heard it.

What comes next

This lesson deliberately stops at boundary behaviour, because each of the things that follow from it now has a lesson of its own:

  • The five kinds of sound field, and why every acoustic formula you will meet assumes one of them — the next lesson.
  • Standing waves and room modes — the predictable frequencies where a room reinforces itself, calculated from a tape measure.
  • Comb filtering in rooms — what happens when a microphone receives the same sound twice and cannot tell.
  • Isolation, absorbers and diffusers — the treatment lessons, where the size rule from this lesson does most of the work.
  • Direct and reverberant fields, and critical distance — covered in the reverb course, since it is the same physics viewed from the effects side.
Studio Rule

Work out the wavelength first. A surface only affects a sound whose wavelength is smaller than the surface, and that single rule explains why bass is hard.

What to practice

  • Work out the wavelength of the lowest note your monitors can produce, then measure the largest acoustic panel you own. If the panel is smaller than a quarter of that wavelength, it is doing nothing at that frequency.
  • Stand outside a venue with music playing inside. You will hear the bass clearly and almost none of the vocal — that is diffraction and absorption, both sorting by wavelength.
  • Clap once facing a bare wall from two metres away, then hold a cushion against the wall and clap again. The difference is one reflection, and it is bigger than you expect.
  • Find a concave surface in your space — a curved window, a domed lampshade, an arched alcove — and listen for the spot where it focuses sound. Then avoid putting a microphone there.
  • Listen to a distant sound at midday and again late at night. It carries much further at night, and the reason is refraction rather than background noise.
Paid Mentor Access
Ask About This Lesson

Students pay for getting unstuck: ask a concept question, routing issue, DAW confusion, or mix decision tied to this lesson.

0
Credits
Unlock direct answers

Free readers can learn from the public Q&A archive. Paid students can ask their own lesson-specific questions and get mentor replies.

Uses 1 credit.
Question saved — a mentor will post a reply here once it's answered.
Answered Questions