Mix a track in a small room, take it to the car, and the bass is either gone or twice as loud as you remember. The usual reaction is to blame the monitors. In most home studios the monitors are not the problem. The room is, and specifically the distance between its parallel surfaces. Before you play a single note you can calculate which bass frequencies your room will exaggerate and which ones it will swallow, because those frequencies fall straight out of the room's dimensions.
Standing waves are just resonance with walls
A room mode is a resonance. When a sound wave's half wavelength fits exactly between two parallel surfaces, the reflection arrives back in phase with the source and reinforces it. The wave stops travelling and stands still in space, with fixed loud spots (antinodes) and fixed quiet spots (nodes). Because the room's geometry is fixed, so is the frequency at which this happens.
For the simplest case, a wave bouncing between two parallel walls, the formula is:
f_n = n * c / (2 * L)Here c is the speed of sound (343 m/s at around 20 C), L is the wall to wall distance in metres, and n counts the harmonics (1, 2, 3 and so on). A 3.4 m wall resonates at 50, 100 and 150 Hz. A 5.18 m wall (17 ft) resonates at 33, 66 and 99 Hz. Every pair of parallel surfaces in the room, length, width and height, produces its own series. When two of those series land on or near the same frequency, the modes stack and that note gets even louder.
The audible result is large. At a pressure antinode an axial mode commonly sits 10 to 15 dB above the surrounding response. At a node the same frequency can almost disappear. That is why a bass note booms at the mix seat and vanishes when you stand up, and why the low end feels completely different in different parts of the room.
Three kinds of mode, in order of damage
Modes are grouped by how many surfaces the wave bounces between. Axial modes use two parallel surfaces and carry the most energy. Tangential modes use four surfaces and run at roughly half the energy, about 3 dB down. Oblique modes use all six surfaces, are weaker still at around 6 dB down, and are usually ignored in basic treatment. The takeaway is simple: deal with axial modes first, because they do most of the harm.
The Schroeder frequency: where the room stops being the problem
Modes do not continue up the spectrum forever. As frequency rises, they become so closely spaced that they overlap and blur together, and the room stops behaving like a set of discrete resonators and starts behaving like a diffuse, statistical field. The crossover point is the Schroeder (or transition) frequency:
Fs = 2000 * sqrt(T60 / V)T60 is the reverberation time in seconds and V is the room volume in cubic metres. For a typical project studio of roughly 30 to 80 m3, Fs lands somewhere around 100 to 200 Hz. That number is the whole point. Almost the entire bass region of a home studio sits below the Schroeder frequency, in the modal zone, where the response is dictated by room geometry rather than by the loudspeaker. Above Fs, ordinary porous panels (foam, mineral wool) tame reflections. Below it, broadband foam does very little, and you are dealing with physics that thin material cannot touch.
Three problems people confuse
Low end trouble in a small room comes from three separate mechanisms, and treating them as one thing leads to wasted money.
- Room modes, described above: dimension based resonances, fixed in frequency.
- SBIR (Speaker Boundary Interference Response): the speaker's direct sound meets its own reflection off a nearby wall. The round trip adds half a wavelength and flips the phase, so the first cancellation notch lands at a quarter wavelength:
fc = c / (4 * d)with d the distance from the speaker to that boundary. Rear wall nulls of around 30 dB are common. For typical two way monitors, a speaker sitting roughly 1 to 2.2 m from the wall puts that notch right in the bass. Closer than about 0.8 m, or well beyond 2.2 m, moves it out of the worst region.
- The listening position null: sitting dead centre between front and back walls, the 50% point, places your head at the node of the fundamental length mode. That frequency can drop by 15 to 20 dB at exactly the spot that feels most natural to sit in.
What actually helps, in order
Position before treatment. Move the listening seat to roughly 38% of the room length from the front wall, centred between the side walls. This is the 38% rule from studio designer Wes Lachot, who looked for points in a rectangular room where standing wave interference is at a minimum. Treat it as a starting point, not a guarantee: it assumes a rectangular room and good symmetry, so verify it by measurement afterward. The one thing to avoid is the dead centre 50% spot.
Pull the monitors off the front wall to move the SBIR notch out of the critical band, and keep the speaker to listener triangle symmetric so the left and right sides behave the same way.
Find the problem cheaply before spending on treatment. Play a slow low frequency sine sweep or filtered pink noise and walk the room, including putting your head into the corners. Where the bass booms loudest is where modal energy collects, and that is where traps belong. Free software such as Room EQ Wizard with any USB measurement mic turns this into a graph and shows the peaks, nulls and decay per frequency.
Then treat. Stack porous bass traps floor to ceiling in the corners first, because corners hold the highest pressure for every axial and tangential mode at once. Depth matters more than anything: 15 to 25 cm of rockwool at 40 to 80 kg/m3 is the cost effective sweet spot, and performs close to commercial studio panels. Thin foam does essentially nothing below about 300 Hz and is the most common wasted purchase in a home studio. Four floor to ceiling corner traps can pull peaks down by roughly 10 dB in a room under about 18 m2.
The limit nobody mentions
EQ and room correction software have a hard boundary that beginners run into constantly. EQ can pull down a modal peak at one point in space. It cannot fill a null. Adding gain at a frequency the room is actively cancelling just burns headroom and changes nothing, because you are fighting a cancellation, not a missing signal. Room correction helps above the modal region and with broad tonal tilt, but the slow decay (ringing) of a mode is a time domain problem that level based EQ does not solve. Physics first, DSP second.
The encouraging part is how much you can predict for free. Divide 343 by twice each room dimension in metres, and you have the fundamental boom frequency for that pair of walls before you have hung a single panel.