Why do transformers hum?

Thread Starter

Sparky49

Joined Jul 16, 2011
833
Hi all,

Electricity is supposed to be our 'silent serveant', so why is it why I walk past a substation or large transformers, I hear a rather loud hum?

I'm guessing that this isn't the movement of the electrons, but some other side-effect.

Can anyone explain this please?:confused:

Sparky
 

chrisw1990

Joined Oct 22, 2011
551
not sure about power stations but i know if you hear high pitched noises from electronics its because the switched mode power supply hasnt got the correct inductor on it, and its something to do with the expansion of the inductor and contraction causing the high pitched noise.. so i was told..
maybe its the same kinda thing with substations..?
 

ErnieM

Joined Apr 24, 2011
8,377
Those large transformers are using a changing magnetic field to transfer a large amount of power. As a side effect that field can wiggle other pieces of metal: the housing, internal lamination layers, whatever it can.

You hear a low pitch as the AC line frequency is low (some 50 or 60 Hz depending on where you are).
 

wmodavis

Joined Oct 23, 2010
739
Electricity may be the silent partner but not necessarily what it does is silent. Think of a buzzer, speaker, motor, etc. All make some sound. Yes the electrons do not in themselves make a sound but can cause sound by their action.

What ErnieM said.
 

Adjuster

Joined Dec 26, 2010
2,148
The fundamental frequencies of 50Hz or 60Hz are usually not dominant in the noises you hear. Typically the noise is mainly 100 or 120Hz, or other higher multiples (harmonics) of the fundamental, either present because of distortion in the original electrical wave, or caused by non-linear processes which generate vibration.

For instance, in magnetostriction the positive and negative current swings often have a similar mechanical effect, chiefly generating vibration at twice the fundamental and some higher harmonics, but little at the fundamental.
 

Adjuster

Joined Dec 26, 2010
2,148
not sure about power stations but i know if you hear high pitched noises from electronics its because the switched mode power supply hasnt got the correct inductor on it, and its something to do with the expansion of the inductor and contraction causing the high pitched noise.. so i was told..
maybe its the same kinda thing with substations..?
A switched-mode power supply might be extra noisy (acoustically) because it has an inappropriate inductor, but if the waveform of current in the power supply has content at audible frequencies for any reason you may hear something.

Mostly nowadays the frequencies are too high to hear, unless irregularities e.g. due to unstable control loop operation introduce lower frequency content. Such effects may sometimes have other causes than incorrect inductors.
 

audio76

Joined Aug 30, 2014
4
Because the magnetic flux density is strongest twice every electrical cycle, the fundamental "hum" frequency will be twice the electrical frequency.
http://en.wikipedia.org/wiki/Mains_hum

Hole Of Knowledge:

A co-worker downloaded a guitar/bass tuner app onto his phone. So, we thought it would be a neat idea to verify what the hum frequency was coming from the shop's transformer. I was expecting to see 6o Hz. What we saw was a reading of 120 Hz (showed up as a "B" note that was a few cents under an actual "B" but not enough to make it a "Bb"). Well, a bit of searching was done and the above quote keeps coming up.

What I don't understand is what do the peaks have to do with the doubling of the frequency? I mean the peaks of a sine wave for example, are referring to the amplitude, are they not? Amplitude doesn't really have anything to do with varying the frequency. I could see that if by raising the amplitude you might cause the harmonic overtones to resonate more sympathetically but I don't see how having "two peaks" has anything to do with it. Although, for a wave to be complete it has to go through a complete cycle which would have to peaks. Again, when talking about frequency, peaks refer to the wave's y-axis (amplitude, volume, voltage...) not the x-axis (degree, frequency...). So, again I don't get how "two peaks" has anything to do with the doubling of frequency.

I'm thinking its the first overtone above the fundamental is what we are hearing not the actual 60 Hz. But Adjuster covered that in this thread already.

If anybody can show me the relationship of having "two peaks" with the doubling of frequency please let me know. The quote is troubling me. I don't see the connection.
 

DerStrom8

Joined Feb 20, 2011
2,390
As mentioned before, transformers operate using an electromagnetic field that forms and collapses at the mains frequency (50/60 Hz). This magnetic field acts on the laminations of the transformer, causing them to move with the field. When the pieces of the transformer move at the same frequency as the current flowing through the windings, it will vibrate at the same frequency. This is what you hear.

Properly mounted transformers will not make much noise at all, and that's why mounting the transformer on rubber pads practically eliminates it. It's the vibrations that you hear, not the electricity.
 

audio76

Joined Aug 30, 2014
4
As mentioned before... When the pieces of the transformer move at the same frequency as the current flowing through the windings, it will vibrate at the same frequency. This is what you hear.
So, you mean that we are hearing the resonating of the actual physical components of the transformer? Like when a group of soldiers walk across a bridge in a synchronize fashion the bridge will oscillate? If thats the case, that would make sense to me. That would mean its resonating right?

But its the "two peaks" part in the wiki quote that doesn't fit right with me. How would having two peaks double the frequency? I'm missing something here....
 

DerStrom8

Joined Feb 20, 2011
2,390
So, you mean that we are hearing the resonating of the actual physical components of the transformer? Like when a group of soldiers walk across a bridge in a synchronize fashion the bridge will oscillate? If thats the case, that would make sense to me. That would mean its resonating right?

But its the "two peaks" part in the wiki quote that doesn't fit right with me. How would having two peaks double the frequency? I'm missing something here....
"Resonating" really isn't the right word, since resonance suggests a capacitor and an inductor that oscillate at a natural frequency. However, yes--it is the entire transformer vibrating that causes the hum. The "two peaks" refer to a positive max and a negative max. The positive max creates one EM field that pulls the core in one direction, and the negative max creates another EM field that pulls the core in the opposite direction. Going from the zero (resting) position at 0v to one of the sides and back to zero happens 100/120 times per second. However, one full cycle (from zero, to one side, to zero, to the other side, back to zero) happens in 1/60th of a second. #12 is right, in most cases what you actually hear is not 60Hz, but 120Hz.

If you want to hear the difference between 60 and 120 (or 50 and 100) Hz, you can use this online tone generator: http://onlinetonegenerator.com/

Simply type in the frequency and hit play. I recommend using the sawtooth waveform due to its clarity over the other waveforms.
 

audio76

Joined Aug 30, 2014
4
"...The "two peaks" refer to a positive max and a negative max. The positive max creates one EM field that pulls the core in one direction, and the negative max creates another EM field that pulls the core in the opposite direction. Going from the zero (resting) position at 0v to one of the sides and back to zero happens 100/120 times per second. However, one full cycle (from zero, to one side, to zero, to the other side, back to zero) happens in 1/60th of a second. #12 is right, in most cases what you actually hear is not 60Hz, but 120Hz.
Magnetostrictive materials can convert magnetic energy into kinetic energy, or the reverse, and are used to build actuators and sensors. The property can be quantified by the magnetostrictive coefficient, L, which is the fractional change in length as the magnetization of the material increases from zero to the saturation value. The effect is responsible for the familiar "electric hum" ( Listen ) which can be heard near transformersand high power electrical devices.
http://en.m.wikipedia.org/wiki/Magnetostriction

"... convert magnetic energy into kinetic energy..."
-->So, one cycle (in this case 60 Hz) will cause a change in the magnetic flux. The cycle no matter what frequency will always produce two movements of the magnetic flux?

"The positive max creates one EM field that pulls the core in one direction, and the negative max creates another EM field that pulls the core in the opposite direction"
-->The change in the field happens twice therefore the core material (the actual stacked sheets within the core...?) are being moved twice which produces an audible clicking noise? Two clicks per cycle?

Therefore:

What we are hearing is the movement of the core as a result of the magnetic flux that is being generated by the 60 Hz AC. Since one cycle produces two EM pulls of the core material you will get two "clicks" every cycle.

So the audible sound isn't resonating sympathetically but being directly generated by the EM pushing and pulling.

Do I got this right?
 

crutschow

Joined Mar 14, 2008
34,280
...........................................................
What we are hearing is the movement of the core as a result of the magnetic flux that is being generated by the 60 Hz AC. Since one cycle produces two EM pulls of the core material you will get two "clicks" every cycle.

So the audible sound isn't resonating sympathetically but being directly generated by the EM pushing and pulling.

Do I got this right?
That is correct. A similar effect occurs for the pulsations of light coming from a fluorescent lamp. They also occur at twice the line frequency which you can see if you observe the light on an oscilloscope with a light sensor.
 

faley

Joined Aug 30, 2014
88
This reply is quite simplistic, but it serves the purpose audio76. (In other words, no offense intended.) Imagine, if you will, that you are standing in a hallway that's not much more than "shoulder" wide. You take your right hand, arm extended to the front and center, then begin swinging your arm to the right, hitting the wall and returning to center each time. Do this once a second. Next, using the same arm, swing your arm to the right, hit the wall, then swing all the way to the left and hit the other wall before returning to center. Do this one time per second. The first is the equivalent of a 1hz single peak (aka "half-wave"). The latter is the equivalent of a 1hz double peak (aka "full-wave").


As I said, it's simplistic but sometimes "simple" works.

Hope this helps audio76.
 

faley

Joined Aug 30, 2014
88
For what it's worth audio76, in my experience, transformer hum is usually the product of loose lamenations; secondary causes are typically loose mounting points.
 

#12

Joined Nov 30, 2010
18,224
True. Millions of small transformers do their job quietly. Some of them hum because they were not properly saturated with the gunk that glues the laminations together. It seems big (kilowatt) transformers always hum loud enough to hear. Explaining this is always about how to communicate with this particular person. Many voices here, and we usually find one that speaks well to the questioner.
 

crutschow

Joined Mar 14, 2008
34,280
The hum of large transformers generally seems to be a fundamental low frequency without the high frequency overtones that loose laminations usually generate. From that I conclude that the noise is simply from the simultaneous contraction and expansion of the all the laminations at twice the mains frequency due to the magnetic field without any significant relative motion between the laminations.
 
Top