This ceiling speaker calculator is free to use, with no account and no password. It runs entirely in your browser; enter your name and email to reveal your result, and we can send you a copy.
The tool answers the three questions a ceiling speaker layout always comes down to: how many speakers the room needs, where they sit on the ceiling, and what transformer tap each one should be set to. Enter the room, the listening plane and your speaker’s dispersion angle and sensitivity, choose how tightly the coverage cones should overlap, and it returns a full layout with plan and elevation drawings.
It is built for the stage where a reflected ceiling plan exists and a speaker has been shortlisted, but the layout has not been set out. It works in metric or imperial units and it produces the numbers that go on a drawing: actual column and row pitch, and the offset from the walls to the first speaker.
There is a step by step walkthrough of every field in the ceiling speaker calculator guide.
A ceiling speaker radiates into a cone. Where that cone meets the listening plane it forms a circle, and the layout problem is covering a rectangle with those circles.
The calculation begins with the throw distance, which is the ceiling height minus the ear height, not the ceiling height itself. That distinction matters more than it looks: a 3.2 m ceiling with seated listeners at 1.2 m gives a 2 m throw, and the same room used standing gives only 1.5 m, which shrinks every coverage circle by a quarter.
The coverage radius is then the throw distance multiplied by the tangent of half the dispersion angle. Doubling the throw doubles the radius and quadruples the area one speaker covers, which is why low ceilings are expensive and tall ones are cheap.
Spacing comes from the coverage diameter multiplied by an overlap factor. Maximum overlap uses 0.50 times the diameter, minimum overlap 0.866, edge-to-edge 1.00 and stretched 1.45. The 0.866 figure is not arbitrary: it is the spacing at which circles on a hexagonal grid meet exactly without leaving gaps, which is why that profile and a hex grid belong together.
Levels are handled separately. The space type sets an ambient noise level, a signal to noise target is added to it to get the minimum needed level, and the calculator works out the power each speaker must deliver over the throw distance to reach it, then picks the nearest standard 100 V line tap from 0.5 W up to 64 W.
Dispersion angle is the most influential number in the calculation. It sets the width of every circle, so a narrower speaker multiplies the count. It also has a frequency dependence that catches people out: manufacturers publish dispersion at several frequencies, and coverage must hold at the highest one your content needs. Paging at roughly 2 to 3 kHz is the widest case, music at roughly 10 to 12 kHz is the narrowest.
Ceiling height and ear height together set the throw. Because coverage area scales with the square of the throw, a modest change in ceiling height moves the speaker count considerably. The tool advises calculating mixed-use rooms at standing height, so the smaller circle governs.
The overlap profile decides coverage quality and cost at the same time, and it is a genuine engineering choice rather than a preference. Stretched at 1.45 times the diameter is correct for background music and paging and is much cheaper. For speech intelligibility it leaves real gaps, and the tool will say so in its verdict.
Grid geometry changes how efficiently circles pack. A hex staggered grid offsets alternate rows and covers a given area with fewer speakers than a square grid at the same profile. Square is easier to coordinate with a tile grid and with services above the ceiling.
Sensitivity and space type do not affect the count at all. They decide the electrical side: how much power each speaker needs and which tap it should be set to.
The built-in example is an 18 by 12 by 3.2 metre room, listened to seated at 1.2 m ear height, with a 100 degree speaker at 89 dB sensitivity, minimum overlap on a hex staggered grid, and Quiet office, classroom as the space type.
Loading it populates the inputs only. You still press Generate result, which is where the tool asks who you are and then returns the layout.
Those inputs give a 2 m throw. At 100 degrees the coverage radius is the throw multiplied by the tangent of 50 degrees, so each speaker covers a circle a little under 5 m across at the listening plane. Minimum overlap then sets a target spacing of 0.866 times that diameter, and the calculator fits a whole number of speakers into 18 by 12 m at approximately that pitch, staggering alternate rows.
Changing one input at a time is the most useful way to use the tool. Switch the listening plane to standing and watch the count rise as the throw shortens. Switch the profile to stretched and watch it fall, then read the verdict to see what that traded away.
The result splits into a layout and a level check, and both matter.
On the layout side, the headline is the speaker count and the grid it implies. Target spacing is the ideal centre to centre distance, but actual column pitch and actual row pitch are what the room resolves to once whole speakers have been fitted, and those are the dimensions that belong on the drawing. First speaker from walls gives the setting out offset so the ceiling can be marked directly. Worst point between speakers is the least well covered point in the pattern, and comparing it against the coverage radius is what produces the verdict.
That verdict is the fastest read on the panel. Coverage complete means every point between speakers falls inside a circle. Cones touch on axis means the circles meet along rows and columns while small diamond areas between four speakers fall outside, which is expected behaviour for that profile rather than a fault. Gaps in coverage means the pitch genuinely exceeds the coverage diameter. Layout impractical, Power short and Above max target each point at a specific input that needs revisiting.
On the level side, the calculator reports the minimum needed level for the space, what one speaker achieves at the ear, the distance loss over the throw, and the headroom remaining below the maximum target. One qualifier travels with those numbers and should not be skipped: overlapping speakers add roughly 3 dB per doubling of contributors, so a tightly packed layout sits above the single-speaker figure shown. The level model assumes free field, with no reflections and no reverberation.
Plan and elevation drawings render alongside, showing coverage circles and speaker positions at ear height, and the dispersion cone in section.
The tool describes itself as rule-of-thumb estimation for system designers, estimators and contractors, and that is an accurate description of its scope.
Its geometry is free field. It does not model reflections, reverberation or absorption. In a carpeted office with an acoustic tile ceiling that is a fair approximation. In a hard-surfaced atrium, a sports hall or an unlined warehouse it is not, because in those rooms reverberation rather than coverage is what limits intelligibility, and a layout that passes here can still be unusable.
It also assumes a single rectangle, one ceiling height and one uniform speaker type. Sloped or stepped ceilings, mezzanines, structural interruptions, mixed speaker models and zoned paging all sit outside the model. And because it works to a coverage criterion rather than a speech transmission index, it cannot answer the question a specification usually asks, which is whether announcements will actually be intelligible.
Reverberation time is the companion question, and the RT60 calculator handles it. For systems using distributed microphones, the PAG/NAG calculator covers feedback stability. Once a design carries real weight, 3D SPL mapping, direct-to-reverberant ratio and STI modelling using manufacturer polar data is the next step, and ALTA Integra takes it from there.
Cookies keep this site working. Analytics show us which pages are useful so we can improve them, and they stay off unless you allow them. We run no advertising here and we do not sell your data. Cookie Policy.
Choose which cookies this site can use. You can change this at any time from the Cookie settings link in the footer.
These cookies are required for the site to work and cannot be turned off. Our hosting provider also runs its own visitor measurement on every page. It is set by the host, not by us, and this site cannot switch it off.
These cookies help us understand how visitors use the site so we can improve it. They stay off until you turn them on.