Subwoofer Enclosure Design: Sealed, Ported and Bandpass, and How to Choose
A subwoofer is half a system; the enclosure is the other half. The same driver in a sealed box and in a ported box of the same volume produces different frequency responses, different maximum output, different transient behaviour and different mechanical risk below tuning. None of that is preference — it follows from the driver's own parameters and the volume of air behind the cone. This article explains what the enclosure is actually doing, what the driver's specification tells you before you cut any material, and how the vehicle changes the answer.

The enclosure exists to solve one problem: at low frequencies a bare driver produces almost no output, because the rear wave wraps around the cone and cancels the front wave. Separating the two is the enclosure's entire job. How it separates them — with a sealed volume of air, with a tuned resonant port, or with both behind a filtering chamber — determines everything else about how the system behaves.
Read the driver before you design the box
Thiele-Small parameters describe a driver's electromechanical behaviour and constrain what enclosures suit it. Three matter most at the design stage.
- Fs — free-air resonance, in hertz. The frequency at which the moving assembly resonates unloaded. It sets the neighbourhood in which the driver naturally works.
- Vas — equivalent compliance volume, in litres. The volume of air with the same springiness as the driver's suspension. A high Vas driver needs a large box to avoid being stiffened by the air behind it.
- Qts — total Q factor, dimensionless. Damping of the resonance. Loosely: below about 0.4 favours a ported alignment, around 0.4 to 0.7 suits either, and above 0.7 points toward sealed or free-air.
Two further figures constrain the outcome physically rather than acoustically. Xmax is the linear one-way excursion — the distance the cone can travel while the voice coil stays fully in the magnetic gap. Beyond it, output stops rising in proportion to input and distortion climbs sharply. Sd, the cone's effective radiating area, multiplied by Xmax gives displaced volume, which is the honest predictor of how much low bass a driver can produce. A large-diameter driver with little excursion and a small one with a lot can displace the same air.
Sealed: the predictable one
A sealed enclosure traps air behind the cone, and that air acts as an additional spring. The result is a gentle low-frequency roll-off of about 12 dB per octave below the system's resonance, with the cleanest transient behaviour of the three types because there is no resonant system storing and releasing energy.
Box volume trades extension against control. A smaller box raises the system resonance and stiffens the suspension — tighter, less deep, and mechanically safer, because the air spring restrains the cone. A larger box lowers resonance and extends response, at the cost of control and of protection against over-excursion at very low frequencies. Sealed enclosures are also the most forgiving of construction error: a volume that lands 10 % away from target shifts the response slightly rather than mistuning a resonance.
Ported: more output, on conditions
A ported — vented, bass-reflex — enclosure adds a tube of a specific length and area. The mass of air in that tube resonates against the compliance of the air in the box, and near that tuning frequency the port radiates in phase with the cone. The system produces substantially more output around tuning for the same input power, typically 3 dB or more, and the driver's excursion actually falls near tuning because the resonant system is doing the work.
The conditions are real. Below tuning the port unloads the cone, excursion rises steeply, and the driver becomes mechanically vulnerable exactly where it is producing least output — which is why a subsonic filter set below tuning is not optional on a ported build. Roll-off below tuning is steeper, around 24 dB per octave. Group delay near tuning is higher, meaning low frequencies arrive slightly later than the rest of the signal. And the port itself has limits: too small an area and air velocity produces audible chuffing and compression at high output, while a larger area demands a longer tube for the same tuning, which has to physically fit.
Bandpass: a filter you cannot adjust
In a bandpass enclosure the driver is not exposed to the cabin at all. It fires into a chamber that vents through a port, so the enclosure itself acts as an acoustic filter. Fourth-order designs seal one side and vent the other; sixth-order designs vent both. Within the passband, output can be high and the driver is mechanically well controlled. Outside it, the enclosure attenuates steeply — including content the system is expected to reproduce.
The engineering caution is that the filter is built from wood. A response error in a sealed box can be equalised; a bandpass passband that is too narrow or centred wrongly cannot be recovered with DSP, because the acoustic energy was never radiated. Bandpass designs also tend to hide distortion products behind the filter, which flatters measurements taken outside the passband and can mask a driver being pushed hard.
| Property | Sealed | Ported | Bandpass (4th order) |
|---|---|---|---|
| Output near tuning | Lowest | High | Highest in passband |
| Roll-off below cutoff | ~12 dB/octave | ~24 dB/octave | Steep both sides |
| Transient behaviour | Best | Moderate; higher group delay near tuning | Poorest |
| Excursion below tuning | Restrained by air spring | Unloaded — subsonic filter required | Restrained by chamber |
| Box volume for a given driver | Smallest | Larger | Largest |
| Tolerance to build error | High | Low — tuning shifts with volume and port length | Lowest |
| Correctable with DSP | Largely | Partly | Barely — the filter is physical |
The car is part of the enclosure
None of the above describes what happens in a vehicle, because a car cabin is small relative to the wavelengths involved. Below the frequency at which the longest cabin dimension equals roughly half a wavelength, the cabin stops behaving as a room with modes and starts behaving as a pressure vessel. Sound pressure rises as the cone displaces air into a sealed volume rather than radiating into free space, and the effect can add 12 dB per octave as frequency falls. This is cabin gain, and it is the reason car systems produce deep bass from enclosures that would sound thin indoors.
Two consequences follow. First, a sealed system with a gentle roll-off often measures nearly flat in-car, because the cabin's rise offsets the enclosure's fall — an alignment that would be bass-shy in a room. Second, a target curve for a car is not flat: measured response typically slopes upward toward the low end, and tuning a car to a literally flat low-frequency response usually sounds thin. Cabin gain also varies with vehicle: a hatchback with the boot open to the cabin behaves differently from a saloon with a sealed boot, and leakage through door seals and vents reduces the effect at the lowest frequencies.
Choosing, in order
- 1Start from the music and the goal. Sustained low content at high level favours ported; accuracy and integration with the front stage favours sealed.
- 2Check the driver's Qts and Vas against the alignment before committing — a high-Qts driver in a ported box usually produces a peaky, one-note result.
- 3Measure the space you actually have, including the volume displaced by the driver, the port and the bracing. Enclosures are almost always designed larger than the vehicle will accept.
- 4For ported, choose tuning from the content, then verify the port fits at an area large enough to avoid chuffing. If it does not fit, the alignment is wrong for the space, not the port.
- 5Fit a subsonic filter below port tuning on any ported build, and verify it with the system playing, not on a bench.
- 6Brace the enclosure. A panel that flexes radiates its own delayed output and subtracts from the driver's, and larger boxes need more bracing rather than less.
- 7Measure in the vehicle, at the listening position, before assuming the design worked. Cabin gain, leakage and boot geometry will have changed the modelled curve.
The design step people skip is the last one. An enclosure is modelled in free space and then installed in a resonant metal box with a hole in it, so the in-car measurement is not a formality — it is the first time the actual system has been characterised at all.
Frequently asked questions
Is a sealed or ported subwoofer box better?
Neither is better in general. A sealed enclosure gives the cleanest transient behaviour, a gentler roll-off and more tolerance to build error. A ported enclosure gives several decibels more output around its tuning frequency for the same power, at the cost of steeper roll-off, higher group delay near tuning and unloaded cone excursion below it.
Why does a ported box need a subsonic filter?
Below the port's tuning frequency the port stops loading the cone, so excursion rises steeply while acoustic output falls. The driver is then working hardest exactly where it produces least, which is a mechanical failure mode rather than a tonal one. A subsonic filter set below tuning removes the content that causes it.
What is cabin gain?
Below the frequency where the cabin's longest dimension is around half a wavelength, a car interior behaves as a pressure vessel rather than a room. Sound pressure then rises as frequency falls — up to about 12 dB per octave — which is why small enclosures produce deep bass in cars and why an in-car target curve slopes upward at the low end.
Does a bigger box always mean more bass?
It means lower extension, not more output overall. A larger sealed box lowers system resonance and extends response downward while reducing control and offering the cone less mechanical protection at very low frequencies. Beyond the driver's suitable range, extra volume mostly costs space.
Can DSP fix a badly designed enclosure?
Only partly. Equalisation can shape response the system is already radiating, but it cannot recover energy the enclosure never produced, and boosting a region where the driver is already at its excursion limit adds distortion rather than output. Bandpass designs are the least correctable, because their filtering is physical.

