The Many Causes of the Low-Frequency Hum

An Engineering Analysis of ELF Vibrational and Acoustical Phenomena
by James Finn © Copyright 2026-02-23

Abstract

For years, we have received reports from all corners of the United States of persistent low-frequency hums in residential and outdoor environments. These hums are frequently attributed to electrical or electromagnetic sources. In practice, low-frequency hum phenomena arise from a range of mechanical, electrical, structural, and environmental mechanisms. This paper provides a systematic engineering analysis of potential causes within the extremely low frequency (ELF) range, including power system magnetics, harmonic resonance, structural coupling, environmental vibration, and perceptual factors. A diagnostic framework for separating magnetic, acoustical, and vibrational mechanisms is presented.

I. Introduction

Low-frequency hums, typically perceived in the 10–200 Hz range, are commonly reported within residential structures but may also be observed outdoors. Accurate diagnosis requires disciplined measurement and distinction between electromagnetic, acoustical, and mechanical sources [1], [2].

II. Power System Magnetic Sources

Transformer core magnetostriction produces vibration at twice the line frequency under sinusoidal excitation, commonly 120 Hz in 60 Hz systems [3]. Harmonic currents from nonlinear loads accumulate in neutral conductors and may induce structural coupling [4].

III. Mechanical and Structural Sources

HVAC systems, compressors, pumps, and rotating machinery operating in the 20–80 Hz range transmit vibration through framing and soil [5]. Ground-borne vibration from rail and traffic infrastructure is also a documented contributor to low-frequency noise [6].

IV. Electrical Coupling and Grounding Effects

Improper bonding and parallel neutral paths may allow circulating currents through structural steel and plumbing [4]. Stray voltage conditions and return currents through earth have been documented in utility environments [7].

V. Utility and Grid-Level Contributions

Large substation transformers generate continuous acoustic emissions at line-related frequencies [3]. Geomagnetically induced currents can increase transformer saturation and associated vibration under rare solar storm conditions [8].

VI. Diagnostic Framework

Diagnosis should include broadband acoustic spectrum analysis, structural vibration measurement, DC and low-frequency magnetic field measurement, and harmonic current analysis consistent with IEC and IEEE measurement standards [1], [4], [9].

VII. Conclusion

The low-frequency hum represents a class of phenomena arising from diverse mechanical, electrical, and environmental mechanisms. Systematic engineering methodology and adherence to recognized standards are essential for accurate diagnosis.

VIII. Testing

 Elexana utilizes state-of-the-art acoustic and vibration equipment to test and identify various types of vibration. We test for the placement of fine electromagnetic equipment such as linear accelerators and electron scanners. 

IX. Solution Notes for Ground-Borne Vibration

• Vibration, like sound, is typically attenuated by “dampening” using absorption or isolation, not shielding. Shielding would also vibrate. You cannot dampen your body.

• Disconnecting your home from the grid would not disconnect you from ground-borne vibration, such as transformer or other equipment vibrations traveling through the soil to your building’s slab or foundation.  

• If the hum is sourced or identified outside the property and we know the direction it is entering from, an isolation barrier or trench could attenuate it. 

Here is a plausible solution, if the source is verified as uncontrolled exposure:

For a 90 Hz ground-borne hum, for example, trench depth is set by the surface-wave wavelength in your soil.

1) How to estimate the wavelength:

For the vibration that usually matters in buildings (Rayleigh surface waves):

λR = vR/f

Typical near-surface Rayleigh wave speeds vR are often 100–250 m/s (which varies widely with soil type, moisture, fill, and rock).

So for f = 90 Hz:

• If vR = 100 m/s: λR ≈ 1.11 m ≈ 3.6 ft

• If vR = 150 m/s: λR ≈ 1.67 m ≈ 5.5 ft

• If vR = 250 m/s: λR ≈ 2.78 m ≈ 9.1 ft

2) Rule of thumb for trench depth

A vibration “cutoff” trench typically needs depth on the order of a meaningful fraction of the wavelength:

• Noticeable attenuation:
  d ≈ 0.3λR (often modest improvement)

• Strong attenuation (typical design target):
  d ≈ 0.5 to 0.7λR

3) What does that mean in feet for 90 Hz

Using the vR range above:

• 0.3λ depth: ~1–3 ft

• 0.5–0.7λ depth: ~3–6.5 ft

• On stiffer ground (higher), it can reach 6–10 ft.

References

[1] IEEE Std 644-2019, "IEEE Standard Procedures for Measurement of Power Frequency Electric and Magnetic Fields."

[2] IEC 61786-1:2013, "Measurement of DC magnetic, AC magnetic and AC electric fields from 1 Hz to 100 kHz."

[3] IEEE Std C57.12.90-2021, "Standard Test Code for Liquid-Immersed Distribution, Power, and Regulating Transformers."

[4] IEEE Std 519-2022, "IEEE Recommended Practice and Requirements for Harmonic Control in Electric Power Systems."

[5] ISO 2631-1:1997, "Evaluation of Human Exposure to Whole-Body Vibration."

[6] FTA-VA-90-1003-06, "Transit Noise and Vibration Impact Assessment." Federal Transit Administration.

[7] IEEE Std 142-2007, "IEEE Green Book – Grounding of Industrial and Commercial Power Systems."

[8] IEEE Std C57.163-2015, "Guide for Establishing Power Transformer Capability while under Geomagnetic Disturbances."

[9] AAPM Report No. 24, "Magnetic Shielding for Medical Facilities," American Association of Physicists in Medicine.