#00215
Replace single-number dB(A) ordinances with limits that also constrain C-weighted and per-octave-band levels at the nearest residence, set as a permit condition rather than a complaint threshold. Prince William County proposed exactly this after observing a 90 Hz tone at a specif
Parent issue
#00205 Continuous low-frequency noise from cooling plant reaches homes at levels ordinances are not written to measure
Tags
Location
Description
A local authority writes three things into the ordinance and then into each permit: a C-weighted limit alongside the A-weighted one, per-octave-band limits covering the 31.5–250 Hz range where cooling plant tones sit, and a measurement point at the nearest residential property line rather than the facility boundary. Compliance is a continuing permit condition with a defined remedy, not a threshold that triggers a complaint process after the fact.
The mechanism of harm is tonality and persistence, not average loudness. A-weighting was derived from equal-loudness contours for quiet broadband sound and systematically discounts low frequencies, so a facility can pass a 55 dB(A) test while producing a hum that is clearly audible indoors at night. Adding dBC captures low-frequency energy; octave-band limits catch a discrete tone that broadband metrics average away.
This is not speculative drafting. Prince William County's March 2025 proposal adds exactly these measures, stating that "the A-weighted scale doesn't effectively address the low-frequency noise generated by Data Centers", citing an observed 90 Hz tone at a specific facility, and proposing residential intermittent limits of 48 dBA day and 43 dBA night (Prince William County).
Two ordinance defects should be fixed at the same time: exemptions for air conditioning equipment (which some ordinances apply to industrial chiller arrays), and the general design assumption that noise complaints concern intermittent events rather than continuous plant.
This requires no new state or national authority. A county or municipality commissions a baseline acoustic survey establishing existing ambient levels at representative receptors, adopts limits expressed as a permitted increase over that measured baseline plus absolute caps, and attaches them to each special exception. The Lyver study is a usable model for the survey stage, having measured 25 receptor locations and modelled buildout (Lyver). Removing by-right development—as Loudoun County did in March 2025—is what makes permit conditions attachable at all (Holland and Knight).
Setbacks are the complement, not the substitute. Sound attenuates with distance regardless of spectrum, so minimum separation from residential zoning does work that no metric alone can.
Links
3Sub-issues
0Case studies
2