11Sound

The short version: Bass has been central to dance-music sound systems from Kingston street dances to Berghain. Two philosophies now coexist: horn-loaded club systems that try to do as little to the signal as possible, and line arrays shaped by digital processing that dominate festival main stages. Signal runs from the DJ mixer to a processor, amplifiers and speakers, and a limiter at the end means pushing the mixer into the red makes things worse, not louder.

A short history of the dance-music PA

Sound systems became popular in 1940s Kingston: trucks loaded with generators, turntables and big speakers for street parties, with custom-built rigs from the late 1950s and some systems later boasting 30,000 watts or more of bass¹¹. The scholar Julian Henriques describes the culture's bass as "enveloping, immersive, intense"¹¹.

In New York, the sound engineer Alex Rosner built the system at David Mancuso's Loft and pushed Bozak to build its DJ mixer²⁹⁵¹¹⁰. Richard Long started out repairing systems for Rosner, went on to install more than 300 of his own, and in 1977 designed the Paradise Garage system, which he treated as his showroom and kept modifying; its "Levan Horn" bass enclosure was named after the club's DJ, Larry Levan²⁹⁵. Long won Billboard's award for best disco sound design in 1980 and died in 1986²⁹⁵.

In the UK, Turbosound was co-founded in 1978 by Tony Andrews, Tim Isaac and John Newsham; Andrews and Newsham left after its 1992 sale to Harman and formed Funktion-One²⁹⁶. Funktion-One builds high-efficiency horn-loaded speakers rather than processing-heavy line arrays. Andrews: "The whole approach with audio for me is to do as little as possible with the signal"²⁹⁶. Its systems are associated with Berghain and Glastonbury's dance areas, among others²⁹⁶. Void Acoustics, founded in 2002, installed in clubs including DC-10 Ibiza and Tresor Berlin (per its own history)²⁹⁷.

On the touring side, L-Acoustics, founded in France in 1984 by the physicist Christian Heil, introduced V-DOSC in 1992 on what it called Wavefront Sculpture Technology - though Heil did not invent line-array theory itself²⁹⁸. d&b audiotechnik was founded in 1981 near Stuttgart by Jürgen Daubert and Rolf Belz, adopted digital processing in 1997 and introduced its Cardioid Subwoofer Array in 2004²⁹⁹. Meyer Sound, founded in Berkeley in 1979 by John and Helen Meyer, moved the amplifier into the speaker with its self-powered MSL-4 in 1994³⁰⁰.

Line arrays and cardioid subs

Line arrays solve an interference problem. Clustered speaker boxes pointed the same way interfere with each other, producing peaks and dips across the audience (comb filtering). A properly aligned vertical column of boxes behaves more like one even wavefront, reducing hot and dead spots and saving truck space³⁰¹. Our observation, as a broad tendency rather than a rule: horn-loaded systems are common in clubs and processed line arrays on big festival stages, and plenty of venues and festivals use both.

Sub-bass is there partly because it makes people move. A McMaster University study switched inaudible very-low-frequency bass on and off every two minutes during an electronic concert, and found dancing increased by 11.8% when it was on³⁰².

The trouble with bass is that it goes everywhere. At 30 Hz the wavelength is longer than 37 feet, and a single 18-inch subwoofer measured nearly the same in front and behind³⁰³. Cardioid subwoofer arrays fix this with timing. In a gradient array, two boxes sit about a quarter-wavelength apart and the rear one is delayed and polarity-flipped, so sound cancels behind and adds up in front; the price is a few decibels of forward output, some punch, and floor space³⁰³. An end-fire array lines subs up front-to-back and delays the front ones most, with the same goal³⁰⁴. The payoff is less low-end bleed onto the stage and toward neighboring stages³⁰¹ - which, for a DJ booth surrounded by subs, matters a great deal.

The chain, and why redlining doesn't work

A DJ mixer has separate house and booth outputs¹¹³. The house feed goes to a front-of-house console or straight to a loudspeaker processor, which applies the crossover, EQ, delay and limiting, often built into the amplifier itself; L-Acoustics' LA4X, for example, is four channels of 1,000 W with filters, up to a second of delay per channel and real-time speaker protection³⁰⁵. Then amps, then speakers.

That limiter is the reason redlining fails. Because the venue's processor caps the maximum level, pushing the mixer into the red only flattens and distorts the music; stacking the mixer's limiter on top of the house limiter strips out transients and tires ears³⁰⁶¹⁴⁷. The advice from the booth is consistent: watch every stage of gain, raise all of them a little rather than one a lot, and keep headroom - "headroom is your friend", as one festival backline tech put it¹⁴⁷³⁰⁷. There is a physiological trap too: DJs develop a temporary hearing threshold shift over a set, so the room seems to get quieter and the gain creeps up³⁰⁸³⁰⁶.

At a DJ-only show (our reading), there is often no "mix" at front of house in the band sense. Much of the engineer's job is managing overall level and protecting the system, and the booth monitors are the DJ's monitor mix. On bigger shows a system engineer arrives first to measure the room, assemble and tune the PA, and walks the coverage zones during the show³⁰⁹.

How loud is too loud

Weighting matters. A-weighting (dBA) mimics how ears respond at low levels and cuts bass heavily; C-weighting (dBC) mimics hearing at high levels and keeps the bass. At 100 Hz, an A-weighted reading is about 20 dB lower than a C-weighted one, so a 100 dBA limit leaves far more room for bass than a 100 dBC limit³¹⁰. LAeq is an A-weighted average over a stated period.

The WHO standard. The World Health Organization's Global Standard for Safe Listening Venues and Events (March 2022) recommends no more than 100 dB LAeq over 15 minutes, plus live level monitoring, better acoustics and sound systems, free hearing protection, quiet zones, and staff training³¹¹³¹².

Limits by country (from a 2019 working-group snapshot published by Live DMA; check the current law before relying on it)³¹³:

Two observations. A 15-minute average is stricter in practice than a 60-minute one, because a loud peak-time quarter-hour cannot be averaged away. And A-weighted limits understate exactly the bass-heavy exposure dance music produces, which is what C-weighted caps are for.

Workers are covered separately. In the UK, music and entertainment came under the Control of Noise at Work Regulations from April 2008, with the Sound Advice guidance for venues including nightclubs³¹⁵. In the US, OSHA's limit is 90 dBA averaged over eight hours with a 5 dB exchange rate, and hearing conservation starts at 85 dBA; NIOSH recommends a stricter 85 dBA with a 3 dB exchange rate³¹⁴. Under OSHA's rule, 100 dBA allows about two hours³¹⁰.

What it does to DJs. In a 2007 study of 30 DJs in Curitiba, Brazil, club levels measured 93 to 110 dB(A), 27% already had measurable hearing loss at high frequencies before their shift, and 66.7% reported tinnitus right after³⁰⁸. Studies cited in the same paper found tinnitus in 74% of Scottish DJs³⁰⁸.

Earplugs work. In a randomized trial at an Amsterdam festival, temporary hearing loss appeared in 8% of ears with earplugs versus 42% without, and new tinnitus in 12% versus 40%³¹⁶. The DJ Anne Savage, quoted by the HSE, noted that many DJs say they cannot mix in earplugs but can wear them the rest of the night³¹⁵.

Report an error in this section