PROBLEMS WITH WIND ENERGY, WIND TUBINS

How to Measure Low-Frequency Noise from Wind Turbines in Homes: A Field-Ready Protocol

TL;DR: This post gives you a complete, practical plan to (1) record indoor low-frequency/infrasound (0.5–200 Hz) and structure-borne vibration, (2) quantify amplitude modulation (the “thump”), and (3) link exposure to sleep and physiology using an ABAB curtailment design. It is designed to stand up in peer review and regulatory settings.

Low-frequency Infrasound Amplitude Modulation Sleep & HRV Curtailment Study

Table of Contents

1. Why measure inside homes?
2. Study design at a glance
3. Equipment & specs checklist
4. Sensor placement
5. 8–10 week protocol
6. Acoustic & vibration metrics
7. Health & sleep endpoints
8. Analysis blueprint (pre-register)
9. Reporting standards & QA/QC
10. Team, timeline, practical tips
11. Printable checklist (quick reference)

1) Why measure inside homes?

People don’t live at the property line—they sleep in bedrooms. Very-low-frequency pressure and building vibration can couple indoors, especially at night. Measuring indoors captures what the body actually experiences and lets you correlate exposure with sleep disruption, heart-rate variability (HRV), blood pressure, and symptom relief when turbines are curtailed or residents sleep elsewhere.

2) Study design at a glance

• Design: Case-crossover + curtailment (ABABAB nights). Randomized shutdown/feather windows (e.g., 22:30–02:30) vs matched control nights.
• Blinding: Residents are not told which nights are curtailed to minimize expectation/nocebo effects.
• Add-on: A short away-from-home phase (3–5 nights) tests symptom reversal while the house stays instrumented.

3) Equipment & specs checklist

Domain	What you need	Key specs
Acoustics (pressure)	2× Class-1 SLMs or equivalent front-ends, + 1–2 infrasound microbarometers	24-bit; Z/A/C/G-weighting; 1/3-oct bands down to ~0.5 Hz; WAV ≥512 Hz; noise floor ≤17 dB(A); proper windscreens/porous hoses
Vibration	3× DC-coupled MEMS accelerometers + 24-bit logger	Bandwidth 0–100 Hz; noise density ≤50 µg/√Hz; mount on slab, interior wall, and bed-frame
Meteorology	10 m mast ultrasonic anemometer + barometer/temp/RH	1 Hz logging; sync to acoustics
SCADA	Operator feed: rotor RPM, blade pitch, yaw, on/off/curtailment	Time-synced to sensors (drift < ±1 s/week)
Sleep/Physiology	Actigraphy watches; overnight HRV straps; automated BP; optional salivary cortisol	1-min epochs for sleep; RMSSD for HRV; morning/evening BP

4) Sensor placement

• Indoors (primary): Bedroom mic at head height; accelerometers on bed-frame and slab (plus one interior wall).
• Outdoors (reference): Mic at 1.5–2 m AGL with secondary windscreen (façade or open yard).
• Secondary room: Living-area mic to confirm whole-house coupling.
• Time sync: GPS or NTP; verify drift weekly.

5) 8–10 week protocol

1. Phase A – Baseline (2 weeks): Continuous logging during normal turbine operation.
2. Phase B – Curtailment (6 weeks): ABABAB nights with 2–3 randomized curtailments/week; match control nights by wind and weather.
3. Optional Phase C – Away: Participants sleep away from home 3–5 nights while home sensors keep logging.

6) Acoustic & vibration metrics

• Levels: LAeq, LCeq, LZeq, LGeq (ISO 7196) with 1/3-octaves from 0.5–200 Hz (indoor & outdoor).
• Amplitude Modulation (AM): IOA Reference Method (10-s windows) + cycle-synchronous analysis keyed to rotor RPM.
• Low-frequency guidance check: Compare indoor bands to DIN 45680 night-time criteria.
• Vibration: RMS/peak acceleration (0–80 Hz), crest factor, pressure–vibration coherence, and “thump” event counts/hour.

7) Health & sleep endpoints

• Primary: Sleep efficiency, actigraphy awakenings/WASO; overnight HRV (RMSSD); morning BP.
• Secondary: EMA prompts (headache, dizziness, ear pressure, nausea, anxiety 0–10); PSQI, Epworth, WHO-5.
• Event buttons: Residents can mark felt “thumps” to anchor exposure snippets.

8) Analysis blueprint (pre-register)

1. Exposure contrast check: Confirm curtailment nights reduce indoor LGeq and AM indices (paired tests/mixed models).
2. Within-person models: Case-crossover or mixed effects: Outcome ~ AM (or LGeq; 1–20 Hz bands) + temperature + wind + random subject.
3. Lag structure: 1, 5, and 15-min lags for arousals/HRV dips vs AM peaks.
4. Dose–response: DIN 45680 exceedance vs symptom probability.
5. Sensitivity: Away-from-home nights and unplanned outages as mechanistic checks.

9) Reporting standards & QA/QC

• Standards hooks: Reference methods aligned with IEC 61400-11/-11-2 (acoustics near turbines), IOA AM method, ISO 7196 (G-weighting), DIN 45680 (indoor low-frequency).
• QA/QC: Field calibration (pre/post), mic self-noise checks, windscreen documentation, 10-s analysis windows, and retention of raw WAV + SCADA.
• Transparency: Preregister protocol; publish code and anonymized datasets where possible.

10) Team, timeline, practical tips

• Minimum team: 1 acoustics lead, 1 field tech, 1 data analyst, 1 clinician/IRB liaison.
• Timeline: Setup 1–2 days → Logging 8–10 weeks → Analysis/report 2–4 weeks.
• Operator coordination: Secure randomized curtailment windows in advance. If not possible, exploit outages and tighten matching.
• Blinding matters: Don’t alert residents about curtailment nights.
• What “good evidence” looks like: Indoor LGeq/AM drop during curtailment and sleep/HRV improve on the same nights.

11) Printable checklist (quick reference)

Open checklist

• ✔ Two Class-1 acoustic channels (indoor/outdoor) with Z/A/C/G; 1/3-oct to 0.5 Hz; WAV ≥512 Hz
• ✔ Infrasound microbarometer & windscreens/porous hoses
• ✔ Three DC-coupled accelerometers (slab, wall, bed-frame)
• ✔ 10 m met mast (wind, T/RH, pressure); 1 Hz logging
• ✔ SCADA (RPM, pitch, yaw, on/off/curtail)
• ✔ Actigraphy, overnight HRV, BP; optional cortisol
• ✔ ABAB curtailment schedule (2–3 nights/week); away-from-home phase
• ✔ Metrics: LAeq/LCeq/LZeq/LGeq; 0.5–200 Hz bands; IOA AM; DIN 45680 check; vibration RMS/peak/coherence
• ✔ Pre-registered analysis; raw data retention; calibration logs

---

Author’s note: If you’d like this post as a downloadable PDF, a one-page field card, or a ready-made spreadsheet for data entry, I can generate those too.