enclosurestudio
← haus-design
No driver loaded

Everything starts with the driver. Load one from the library, paste a spec sheet, or type the numbers in. The tool works out the rest of the physics from there.

What kind of speaker is this?

The low-frequency driver sets the box either way. A second driver adds a crossover, and changes what has to fit on the baffle.

Driver library

Parameters taken from manufacturer datasheets. Always confirm against the sheet that came with your actual driver — makers revise them.

Driver parameters

The three that set box size

Get these right and everything downstream is right.

Electrical & mechanical

These let the tool predict loudness, excursion and amplifier needs.

Cone & travel

How much air the driver can physically push.

What the tool worked out

Derived from the parameters above — no extra data needed.

Every enclosure is a trade between size, depth of bass, accuracy and loudness. Tell the tool which of those matters most and it ranks the workable options for your driver.

What matters most for this speaker?

This reorders the recommendations. Nothing is hidden — the maths is the same either way.

Ranked options

Side by side

The chosen design, with the four checks that decide whether it survives contact with real music: how it responds, how far the cone travels, how fast the air moves in the port, and how loud it goes before something gives.

Enclosure

Adjust anything — the plots follow immediately.

Drive conditions

What you actually feed it. Excursion and port noise both scale with this.

Frequency response

How the bass rolls off

Anechoic half-space. The −3 dB point (F3) is where output has fallen to about 70% of the midband — the usual shorthand for "where the bass ends". A room will typically add back several dB below 60 Hz.

Cone excursion vs Xmax

The most important plot here. Cross the red line and the coil leaves the magnet gap: distortion, then mechanical damage. Ported boxes unload badly below tuning — that's the steep rise on the left.

Port air velocity

Air speed in the port. Past roughly 17 m/s it starts to hiss and chuff on bass notes; past 25 m/s it's obvious and the tuning itself starts to shift. Bigger or flared ports fix it.

Maximum SPL — how loud before something gives

Two ceilings at once. The excursion limit is where the cone runs out of travel; the thermal limit is where the voice coil runs out of power handling. The lower of the two is your real ceiling at that frequency.

Numbers

Where you hand over from woofer to tweeter, and what that costs. The component values below are exact for a resistive load — real drivers aren't resistors, so treat them as the design you build and then measure, not the one you ship.

The box as an object. Drag any dimension and the others move to hold the volume — so you can chase a proportion the industrial design needs without ever losing the acoustics. Braces and ports are real geometry here: what you place is what gets subtracted from the air inside.

Dimensions

Internal dimensions. Lock one and it stays put while you drag the others.

Bracing ?

Real panels, not a percentage guess. Volume comes off the box automatically.

Ports

Drag a port on the drawing to move it on the baffle.

Driver on the baffle

Position only — it does not change the acoustics, but it does change what fits.

Net acoustic volume is not the box you build. The magnet, the port tube and the bracing all steal air back. Build to the acoustic figure and the finished box comes out undersized — this page adds it all back.

Material

Sets wall thickness, external size and finished weight.

What else steals air

Everything inside the box that isn't air.
Shape, proportions, bracing and port placement now live on the Cabinet tab, where you can drag them directly.

Volume, step by step

Cut sheet

Panel behaviour

A panel that flexes radiates sound of its own — slightly late, and coloured by its own resonance. You want that resonance well above the driver's working range, and you want it to stop ringing quickly once excited. Figures assume simply-supported edges, which is the pessimistic end; real glued joints land somewhere between this and about 1.8× higher. Bracing a panel across its middle roughly doubles its first mode.

Standing waves inside the box

Each internal dimension rings at its own half-wavelength. Good proportions spread these apart so no single frequency gets reinforced three times. Stuffing and internal absorption knock them down further.

Everything on one card

Speaker design has its own vocabulary, most of it named after the two engineers who formalised it — Neville Thiele and Richard Small. Here's the whole set in ordinary language.

How the tool does the maths

Sealed box. Vb = Vas / ((Qtc/Qts)² − 1) and Fc = Fs · Qtc/Qts. The response is an ordinary 2nd-order high-pass, so F3 has a closed form and no searching is needed.

Ported box. The response comes from the full 4th-order vented transfer function including box leakage, with s⁴ + a₁s³ + a₂s² + a₃s + a₄ where a₁ = 1/Qts + h/QL, a₂ = α + h + 1 + h/(Qts·QL), a₃ = h/Qts + h²/QL, a₄ = h², h = Fb/Fs and α = Vas/Vb. F3 is read numerically off that curve rather than from the usual approximation, which can be a couple of Hz out.

Cone excursion. Derived from real motor parameters: Cms = Vas/(ρc²Sd²), Mms = 1/(ωs²Cms), Bl = √(ωs·Mms·Re/Qes), and peak displacement x₀ = e·Bl/(Re·Mms·ω²). In a ported box the cone displacement carries the numerator s² + (h/QL)s + h², which is why it notches at tuning and runs away below it.

Port. Lv = c²Av/(4π²Fb²Vb) − k·r, with the end correction k set by how the ends are finished. Air velocity is computed from the port's own volume velocity at the drive level you set, not from a worst-case assumption.

Loudness. Reference efficiency η₀ = (4π²/c³)·Fs³·Vas/Qes, sensitivity = 112 + 10·log₁₀(η₀) dB at 1 W / 1 m into half space. Maximum SPL is the lower of the excursion-limited and thermally-limited curves at each frequency.

What it does not model. Baffle step, edge diffraction, room gain and room modes, driver non-linearity, power compression, motor inductance effects above a few hundred Hz, or passive radiators. Real in-room response below 100 Hz is dominated by the room and will differ substantially.