Industrial plant noise is rarely one sound. A boundary usually has to carry a fan stack, a cooling tower, a compressor house, a chiller bank and yard traffic at the same time, and whatever you build has to answer for all of them. That is why the first question on a plant project is not how high, but which source actually reaches the boundary.

Two different jobs, two different products

Most plant noise complaints are solved by one of two things, and they are not interchangeable. Choosing the wrong one is the most common reason a project spends money and still fails the measurement.

The boundary wall

A wall is the right answer when the receiver is beside the plant and the sound is leaving the site roughly horizontally. It works by blocking the direct path and forcing the sound to diffract over the top edge. Because diffraction is the mechanism, the height of the wall relative to the source and the receiver matters more than the panel. A wall that is high but far from the source does less than a lower wall close to it.

The equipment enclosure

An enclosure is the right answer when one machine dominates and the receiver surrounds it. It works by cutting the sound at the source rather than downstream, so it is far more effective per square metre than a boundary wall. The price of that effectiveness is a hard constraint: the machine inside still has to breathe, and an enclosure that ignores airflow will overheat the equipment or trip it on high discharge pressure.

Why low frequency decides the design

Fans, cooling towers and large blowers put most of their energy into the low end of the spectrum. That single fact drives three decisions at once.

First, mass. A thin solid panel resists sound in proportion to its surface mass, and the mass law gives roughly six decibels for every doubling of mass. Doubling the thickness of a steel skin is expensive and gains little compared with adding mass where it is cheap, which is usually the infill and the frame.

Second, height. Low frequency does not travel in a straight line the way high frequency does; it bends over the top edge of a wall and around the ends. This is why a barrier that removes high frequency noise completely can barely move the overall level when the dominant source is a fan. If the reading is dominated by low frequency, adding height to the wall generally does more than upgrading the panel.

Third, absorption. Inside an enclosure, a reflective lining lets sound bounce between the walls and build up around the machine, which reduces the benefit of the enclosure itself. A perforated absorptive cassette on the inside face, with the protective fabric and infill behind it, breaks that build-up. On a boundary wall, absorption matters when there are reflective surfaces on the far side, such as a façade across a narrow road.

What the number actually depends on

No manufacturer can quote a plant barrier from a photo, because the answer depends on five things that only the site can supply:

• The sound power level of each source, and which one dominates at the boundary.

• The distance from source to boundary, and from boundary to the nearest sensitive receiver.

• Receiver height. A first-floor window and a ground-floor window are different problems.

• Ground surface between source and receiver. Hard paving reflects, grass and planting absorb.

• Tone and impulsiveness. A steady hum and a cycling compressor with the same A-weighted level are judged differently, and some limits treat them differently too.

Materials that survive a plant

Plant environments are harder on materials than roads. Heat, oil mist, condensate and vibration all shorten the life of a lightweight panel. The practical specification is a galvanised or powder-coated steel cassette, mineral infill that does not settle, and fixings that can be re-tensioned. Where the enclosure has to pass air, acoustic louvers with a free area sized by the equipment supplier do the work that a solid panel cannot.