Free calculator

Baffle step calculator: transition frequency and compensation

Below a certain frequency a loudspeaker no longer radiates only forward but all around – the on-axis level drops by up to 6 dB. Enter the baffle width: the calculator shows where this baffle step lies and suggests a passive compensation.

Inputs and result

cm
Ω
dB

6 dB is free space. Near a wall or in a room, 3–4 dB usually compensates better.

Centre of the baffle step (f3): 575 Hz

Compensation network: an inductor with a resistor across it, in series before the woofer

PartPositionValue
L1L ∥ R, in series1 mH
R1L ∥ R, in series4.7 Ω

Above the step the woofer plays 4 dB quieter – the tweeter and midrange have to be matched to that (e.g. with an L-pad).

Formulas

f3 ≈ 115 / W   (W in m)
k = 10^(A / 20)   (A = compensation in dB)
R = R_driver · (k − 1)
L = R / (2π · f3 · √k)   → network centred on f3
Simulate in Crover with real measurements

Opens a new project with exactly this circuit. Works as a guest without an account; then load your drivers' FRD/ZMA.

What is the baffle step?

At high frequencies the baffle is large compared with the wavelength: the sound goes forward (half space). At low frequencies it wraps around the cabinet (full space) – on axis up to 6 dB are then missing. The transition is roughly at f3 ≈ 115 / W (W = baffle width in metres).

It is usually compensated with a larger low-pass inductor or a network of its own: an inductor with a resistor across it before the woofer. At low frequencies the inductor passes everything, at high frequencies the resistor limits the level.

Limits of the rule of thumb

  • The formula knows only the width. Driver position, cabinet height and depth and rounded edges shift the transition and cause ripples of a few dB. Centred drivers show the deepest dips.
  • How much to compensate depends on placement: near a wall or in a small room part of the missing 6 dB comes back.
  • The network is calculated for a resistive woofer. With its real impedance (ZMA) and the inductor's DCR the effect changes – best simulate it together with the low-pass inductor.

Sources: the f3 ≈ 115/W rule of thumb after Olson's diffraction measurements (H. F. Olson, “Direct Radiator Loudspeaker Enclosures”, 1951), as used by Rod Elliott (ESP, “Baffle Step Compensation”) and in Jeff Bagby's Baffle Step Designer.