Contents
- Why an enclosure loses its decibels somewhere other than its walls
- Mistake No. 1: ventilation sized last
- Mistake No. 2: leaking doors and penetrations
- Mistake No. 3: rigid connections that make the wall radiate
- Mistake No. 4: performance specified without a spectrum
- Our method: sizing the acoustic enclosure as a whole
Why an enclosure loses its decibels somewhere other than its walls
An enclosure is a system: its performance is set by its weakest point, not by the quality of its panels. As soon as one function is sized in isolation, that function caps the overall result.
The enclosure is one of the most widely used means of reducing noise radiated by a machine. It is also one of the most sensitive to detail. In laboratory tests published in 2001, INRS measured the same double-wall test enclosure before and after fitting a floor seal. Its insertion loss rose from 19 to 45 dB(A) (Cahiers de notes documentaires No. 182, ND 2144).
Same walls, same source: a 26 dB(A) difference, caused by a single seal.
Mistake No. 1: ventilation sized last
An enclosed machine continues to dissipate heat, and every air inlet or outlet is an opening in the enclosure. Left untreated, a single opening can cut acoustic performance in half.
The figures involved add up quickly: removing 10 kW of heat loss with the air temperature rise limited to 10°C already requires around 3,000 m³/h. In the same INRS tests, a single 40 × 20 cm opening dropped the insertion loss of a metal enclosure from 38 to 23 dB(A). The degradation was more pronounced at high frequencies, which the opening lets through almost unimpeded.
Fitted with a ventilation silencer combined with flexible baffles at the inlet and outlet, the same opening brought the overall performance back to 38 dB(A), the level of the closed enclosure.
The solution exists, but it still needs to be sized together with everything else. The silencer adds a pressure drop that the fan must overcome, and the air velocity between its baffles generates its own noise. An under-ventilated enclosure causes the machine to overheat: the temptation to run it with the doors open becomes strong, and its acoustic performance disappears along with them.
Mistake No. 2: leaking doors and penetrations
A tiny leak can dominate the insulation of the entire enclosure. In theory, a gap representing 0.1% of the surface area caps the overall performance at around 30 dB, regardless of wall quality.
The insulation of an enclosure is calculated by weighting the acoustic transparency of each element by its surface area. On another test enclosure, INRS regained an additional 6 dB(A) by treating leaks around, then underneath, a door.
An acoustic door must offer the same properties as the fixed panels, seals included. These seals compress and wear out over time: specifying replacement parts is part of the sizing process. The French labor code even lists maintenance programs among the ways to reduce noise exposure (Article R4434-1).
The same logic applies to every cable, pipe, or shaft penetration: an acoustic sleeve planned for in the specifications, not added as a fix after acceptance.
Mistake No. 3: rigid connections that make the wall radiate
An enclosure treats airborne noise, not noise transmitted through the structure. If the machine touches the enclosure, the wall itself becomes a noise source.
When the machine, its frame, or a pipe touches the enclosure, vibrations excite the walls, which radiate like the membrane of a loudspeaker. The insulation of the panels then plays no role at all: the noise does not pass through the wall, it is produced by it.
These vibrations concentrate on tones linked to the machine’s rotation, most often in the low and mid frequencies. They amplify when they coincide with the natural frequencies of the panels.
The solution: no rigid bridge between whatever vibrates and the enclosure. Machine on elastic mounts, flexible couplings on pipes and ducts, enclosure decoupled from the machine and its frame, damped panels where necessary. Article R4434-1 of the French labor code also cites the reduction of structure-borne noise, through damping or isolation, right after enclosures.
Mistake No. 4: performance specified without a spectrum
An enclosure specified as “25 dB(A)” without a source spectrum is not properly specified. Its performance is read as a curve, frequency by frequency, not as a single figure.
The insulation of a wall increases with its mass and with frequency, then drops around its critical frequency, which must be kept well away from the machine’s emerging tones. At low frequencies, effectiveness also depends on the distance between the machine and the walls: this distance must increase as the lowest frequency to be treated decreases. An overly tight enclosure first loses effectiveness at low frequencies.
The spectrum must cover all operating speeds. On a variable-speed machine, the tones drop with the speed, moving into the range where the enclosure attenuates least.
Acceptance testing is prepared from the specification stage onward. The NF EN ISO 11546-2 standard governs on-site measurement of the insertion loss of a complete enclosure, for acceptance and verification purposes. Defining this method, the measurement points, and the operating speeds in the specifications avoids discovering too late that what is being measured is not what was calculated.
Our method: sizing the acoustic enclosure as a whole
At Groupe BOËT, all four points are addressed together, starting at the pre-sales stage, rather than being fixed after acceptance.
“An enclosure must be sized as a whole: walls, ventilation, access, penetrations, and supports, based on the machine’s actual spectrum and across its full range of operating speeds. In our view, that is the condition for the performance measured at acceptance to match the performance calculated at design.” R&D Team, Groupe BOËT
Our R&D team recommends addressing this from the consultation stage onward. Our expertise in acoustics, airflow engineering, and vibration engineering covers all four points above: wall insulation, the sizing of ventilation silencers and their pressure drop, and the measurement, analysis, and treatment of vibrations.
For you, the stakes are concrete: achieving at acceptance the performance calculated at design. Discover our acoustic enclosures.
Conclusion
Sizing an acoustic enclosure rarely comes down to wall thickness. It comes down to ventilation, the sealing of doors and penetrations, the absence of rigid connections, and accounting for the actual spectrum across all operating speeds. Addressed together from the design stage, these four points ensure that the performance measured at acceptance matches what was calculated. Addressed as an afterthought, they cost decibels and rework.
Have an enclosure to size, or performance to recover? Contact our engineering team.