Working diffusers, not wall art. Every well depth comes from the quadratic residue sequence n² mod N (Schroeder), so a tile spreads a reflection into many directions and arrival times instead of sending it back as one echo. They do not absorb sound and do nothing for bass, room modes or loudness; for a trap tuned to one room mode, see the Bass traps tab. Deep tiles treat a room from about their design frequency up, once the wall is wide enough; block tiles are a shallow 2D diffuser for upper mids and treble. Formulas after T. J. Cox and P. D'Antonio, Acoustic Absorbers and Diffusers, 3rd ed., CRC Press, 2016. The band shown is design intent, calculated from the shape of the tiles, not a measurement. The ready-made tiles for the A1 mini, MK4 / Core One and P1S / X1 are also on Printables.
Seen from the room
What you see on this page, explained
A trap for one room mode. A folded quarter-wave resonator, printed in blocks of 120 x 120 mm cells, each one tuned to a single mode of your room. It is a narrow absorber: it shortens the ringing of that one mode and does nothing for the other modes, the treble or flutter echo. For broadband bass, porous or membrane traps in the corners do more per kilogram of material; this is a narrow tool for one ringing mode. Every number on this page is calculated, not measured: nobody has printed these cells and measured them in a room yet. If you try one, we would like to hear your before and after results in GitHub Discussions.
Your room decides which modes ring.
You rarely know this, and it matters: in the example room eight cells give about 1.5x faster decay with lively or medium walls, and about 1.05x with damped walls.
Room modes up to 120 Hz
Per cell, tangential and oblique modes trapped at a corner or an edge gain most (in the 3D simulation about 1.7x more per cell than an axial mode at 4 cells), but the model is less exact there.
| Frequency | Decay now | Verdict |
|---|
One cell is 120 x 120 mm on the wall and 50 - 150 mm deep. More cells help less and less: the decay levels off at about 1.5-1.75x faster however many cells.
5 mm kept free on every axis.
The ZIP holds the block STLs, the end-cap plate (one cap per cell, needed: it closes the channel) and a README with the print plan.
Optional. Print them only if you can measure the decay before and after; without a measurement use the nominal end caps from the ZIP.