43
$\begingroup$

Often times you pass by an electrical box on an electrical pole and you hear a distinct hum emanating from it. What causes that tone? Does the flow of electricity itself have a sound? Or does the flow rattle the metal parts at a certain frequency, causing the sound?

$\endgroup$
  • 4
    $\begingroup$ There are two mechanisms: magnetic and electrostatic. The "box on the pole" is usually a transformer. The changing magnetic field in the transformer can move the magnetic core and the wires slightly and cause a sound. This is similar to the way a speaker works, except that there is no strong permanent magnet and that transformers are usually tightly wound and impregnated to reduce this effect (being silent is a criterion for a well made transformer). The other mechanism is high voltage electrical discharges in air which can cause a hiss that is modulated by the line frequency. $\endgroup$ – CuriousOne Sep 12 '15 at 19:33
  • $\begingroup$ Just an intuitive guess here, but the AC current generates a changing magnetic field. Maybe that causes the wires to vibrate in Earth's magnetic field. $\endgroup$ – Sponge Bob Sep 12 '15 at 19:33
  • $\begingroup$ Another effect relevant for poles with power cables: Wind might cause noises here due to aeroacoustic effects. $\endgroup$ – Sebastian Riese Sep 13 '15 at 11:56
  • $\begingroup$ I always thought that electric hum was due to ionization of air near exposed conducting areas that are charged, especially with AC. I once diagnosed a failed power supply on a computer by a subtle hum it was making and figured out the frequency as a multiple of the mains frequency. However, my home pc also rattles due to fan and hdd motion interacting with the case (tested by applying pressure to the case and blocking fans) and the sound is distinctly different and mechanical, with different harmonics, AFAIK not multiples of the mains frequency (which is irrelevant since they're DC). $\endgroup$ – mechalynx Sep 13 '15 at 13:11
  • 1
    $\begingroup$ Several people here mentioned transformers. The iron core of a transformer is somewhat magnetostrictive: en.wikipedia.org/wiki/Magnetostriction $\endgroup$ – Solomon Slow Sep 14 '15 at 13:46
51
$\begingroup$

Varying (due to AC) electromagnetic forces exerted on the components cause them to vibrate thereby causing the hum. Components that typically hum noticeably are transformers (where the coils and cores vibrate/magnetostrict in the varying magnetic field), and under certain circumstances capacitors (typically they have higher resonant frequencies and audible capacitor hum is typical for electronic equipment, e.g. computers, here the forces on the plates cause the dielectric to oscillate mechanically).

The frequency of the typical hum is the mains frequency of (in Europe) $50\,\mathrm{Hz}$ or $100\,\mathrm{Hz}$ (if it is due to magnetostriction or there are rectifiers that effectively double the lowest frequency), respectively $60\,\mathrm{Hz}$ ($120\,\mathrm{Hz}$) in the US. In certain cases the most audible frequency might also be a higher harmonic (due to nonlinearity, resonance phenomena and the response of the human ear).

The mechanical oscillations due to magnetostriction double the frequency, because the length change does not depend on the direction of the magnetization (so it will have one extremum when the flux through the transformer core reaches zero and the other when the flux is maximal or minimal).

$\endgroup$
  • 2
    $\begingroup$ And in airplanes, it is 400 Hz. You can sometimes hear this when the plane is at the gate, running on 'shore power' with the engines off. $\endgroup$ – alex.forencich Sep 14 '15 at 8:13
9
$\begingroup$

Switching power supplies make higher-pitched noise, because they rectify to DC and then drive a small transformer at much higher than mains frequency.

It's still the same root cause: magnetic forces vibrating the components, specifically the inductors. I'm not aware of electrostatic attraction/repulsion in capacitors ever being a source of noise. Capacitive microphones and piezo-electric buzzers and mics exist, but they are specifically designed as transceivers, not just capacitors.

So-called "coil whine" is common for computer parts, like motherboards, graphics cards, and main computer power supplies (that the cord plugs in to).

High-power digital logic chips internally use supply voltages between 0.8 and 1.2V these days, at high current. Different components convert the common 12V rail down to the exact voltage they're designed for. (It's not a coincidence that the supply voltage is near the band-gap for silicon. Running at as low a voltage as possible for a given CPU frequency minimizes power usage, reducing heat (which is the limiting factor in making digital logic faster these days)).

$\endgroup$
  • $\begingroup$ One small addition: I am pretty sure that in high frequency applications capacitors make noise as well (because the dielectric can oscillate longitudinal). The structural integrity of ceramic capacitors is certainly not threatened by some vibration. $\endgroup$ – Sebastian Riese Sep 13 '15 at 12:19
  • $\begingroup$ @SebastianRiese: Good point, corrected my wording. I'm not aware of capacitors being an audible noise source in computer components. Inductors are usually mounted sticking up from a circuit board, while small capacitors are often surface-mounted. This gives inductors more ability to move air. $\endgroup$ – Peter Cordes Sep 13 '15 at 12:33
  • $\begingroup$ My intuition says: Even surface mount components can easily vibrate (given you are close enough to an eigenfrequency), and then they make the epoxy vibrate which acts as soundboard. I am sure I read that capacitors make noise somewhere but I can't find the source right now (said source said that in computers what you hear are typically capacitors, not coils). $\endgroup$ – Sebastian Riese Sep 13 '15 at 14:52
  • $\begingroup$ Capacitors are at least doing it the other way around: en.wikipedia.org/wiki/Microphony there it says especially X7R ceramic caps act as microphones because the material is piezoelectric, so this should work both ways. $\endgroup$ – Sebastian Riese Sep 13 '15 at 14:55
  • $\begingroup$ @SebastianRiese: I made an edit to suggest that capacitors which aren't designed as electroacoustic devices are unlikely to interact much. For the shaking-apart thing, I had electrolytic and mylar-thin-film caps in mind. Thanks for the feedback. :) $\endgroup$ – Peter Cordes Sep 14 '15 at 10:57
1
$\begingroup$

The power pole boxes you mention, are typically step down voltage transformers, and although it is true that the source of the "hum" could be the coils and the core, I think it is more likely that the transformer enclosure might be loose, causing it to vibrate at the line frequency (50 or 60 Hz).

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.