UKC

I spend my working week in a workshop making furniture,  his involves using loud machinery.  Until yesterday I wore some ear defenders with Bluetooth speakers in them (3M worktunes) so I could listen to podcasts and music. They have broken!!  This is serious.

They were not noise cancelling, rather just standard industrial ear defenders rated at 27db noise reduction.  So they block noise out passively and play music inside.

I’m thinking that active noise cancelling headphones may work better in blocking sound and have better audio quality, the old ones were not great in terms of music quality.

What I do not understand from a physics standpoint, if the ANC is playing the inverse waveform of ambient noise in order to cancel out the loud machine, is my ear experiencing twice the noise in effect or is it saving my hearing?

I could get noise cancelling earbuds and put these under some passive ear defenders but in my experience it’s not physically comfortable to do that for long periods and annoying to take them on and off.  I’d rather have a single set of headphones.

I could just get more of the 3M worktunes that have lasted a good 4-5 years but I would like better sound and better protection would be good.

Finally any recommendations for NC headphones.  Happy to spend some cash on Bose if they are worth it but they will have a hard life in the workshop so if I can spend less I’ll cry less when they break.

Thanks UKC

Think you might be mixing up headphones with something else, there is a system for cancelling noise, by playing the inverse, but that usually not for headphones.

https://splice.com/blog/how-does-noise-cancelling-work/?utm_source=google&a...

Isn’t that article saying just that, ANC works by listening to the outside noise and playing the inverse wave form to cancel out the 2 phases. 
 

Am I missing something fundamental?

Your ear is not experiencing twice the amplitude. 

I love my ANC headphones but I’d probably stick with ear defenders if I was worried about actually protecting my ears. Maybe some good earphones inside them? 

I've not tried them myself but these keep getting recommended on a few woodworking groups I follow:

https://isotunes.co.uk/collections/hearing-protection 

In the work place you shouldn't be wearing noise protection that will stop you hearing outside noise of 75dB(A). To do with still having the ability to hear alarms, etc.

Not that most employers will bother. 

This technology was developed not for entertainment but for crews in overpoweringly noisy environments (AFVs, battlefield noise, aircraft etc) and it really does work.  It's quite old technology as well except that there is now far more computing/DSP available in a small space.

Some of the ideas are very old indeed, the idea of microphones cancelling noise on the battlefield by having a double microphone, one facing the mouth and the other facing away so that the noise picked up away from the mouth can be fed in antiphase to cancel non-speech noise, eg explosions, gunfire or wind noise.  They really work.

Noise cancelling earphones aren't suitable for your purpose. Get some new 3M.

If you tried cancelling loud noise with imperfect inverse signal you would get errors from the cancelling with amplitude similar to the quiet noise you want to hear. This is probably okay for speech - so the noise cancelling for video conferencing - but won't work for music. Also it just wouldn't be safe for your ears long term.

Being self employed and the only person in the workshop I’m not worried about the ‘rules’ I’d like it as quiet as possible. 
 

Thanks.  I’ve got some iso-tunes ear buds but they are useless at blocking loud sounds.  So you just end up with the ambient noise combined with the music cranked up so you can hear it.   Then they broke anyway. 
 

haven’t tried the over ear style ones but they have a lower db rating than my 3M ones so seem like a retrograde step. 

The physical sound is a series of under- and over-pressure wavefronts travelling through the air and  impinging on your eardrums.

Properly working sound cancellation sends a matching train of wavefronts onto your eardrum, with a low pressure wavefront hitting your eardrum just as an equal and opposite high pressure one from the sound hit, and so on.  So instead of having your eardrum wobbled about by the pressure variations from the noise spruce, it sees none and stays nice and still.  Physically, as far as it’s concerned, the sound isn’t there.

If it was set up wrongly and the pressure variations weren't cancelling but were reinforcing, you'd hear that as a louder sound.

It's interesting to think about the energy being transmitted by each of the noise source and the anti-noise emitter; both are clearly turning some other form of energy in to sound waves; yet your ear experiences no pressure change and so no energy lands on your ear.  Conservation of energy and all that means it must be going somewhere.  For some other locations in space, the waves reinforce each other and the sound is louder - the energy from the source is not absorbed by active antinoise, but rather displaced to somewhere other than your ear.

Now, if one imagines a system running antinoise  for some specific locations, for example as part of some covert anti-surveillance technology, there would be some other areas of displaced high intensity.  It's conceivable such a system might use the audio frequency beat between separate ultrasound beams to inaudibly damp audio frequency vibrations in window glass from a distance, for example.  I certainly hope the designers of any such system would have considered the unwanted nodes of high energy density such a system would produce in other locations...  

https://www.dni.gov/files/ODNI/documents/assessments/2022_02_01_AHI_Executi...

I've tried my bose QC35 noise cancelling headphones at work. They are much more comfortable for all day use than the peltor ear defender/talkback headsets that we use for comms.

Whilst they work well in quieter event settings (polite ambient crowd noise, continuous drone of industrial air con units etc.) as soon as stadium PA systems or loud engine noise is involved the ANC is not man enough to cope and the headphones don't do enough physically to keep the higher db noise out.

In my experience, the loudest sounds my QC35 work well for really is the annoying screechy bits of the London Underground. They're also great on flights at reducing cabin engine noise about a third. Anything louder you'll probably have to crank the tunes up to more hearing damaging levels. Can't say how much better the new improved QC 45s are, but would say look after your lug'oles! 

From the article. The first paragraph is used external to headphones.

However, if we were to move one of the waves out of phase by 180 degrees, we would hear nothing! When one signal is played at an inverted phase, the peaks and troughs of the two identical waves cancel each other out. If two signals cancel each other completely, you can say that they phase cancel, or null, with each other. This is an important technique in many other audio engineering applications as well, but we’ll have to revisit that in another blog post.

So, how does this relate to headphones? Well, active noise cancelling headphones could technically be referred to as headsets since they have microphones embedded within them. The onboard mics record the outside noise that’s meant to be cancelled, and then the headphones’ DSP (digital signal processing) chip analyzes the noise, using the manufacturer’s algorithms to predict the most accurate possible representation of the outside noise that the headphones’ mics are picking up. Then, the noise is removed from the signal via phase cancellation as the headphones invert the phase of the noise and add it to the playback, resulting in less outside noise reaching the listener’s ears.

This isn't true of noise cancelling headsets, the noise is filtered out using DSP in the headphones.

There are noise cancelling systems which work out in the open air, they work best where the noise is pretty constant, compressor houses are a good example, these are the invert wave type.

No. If you put noise cancelling headphones on and turn ANC off they don't block much sound at all, so what would they be filtering the noise from? They actively produce an inverse waveform to cancel external noise out, using the onboard DSP.

You are misunderstanding the bold sentence. Yes, that is a common technique but it is not stating that it is not used for ANC. It is the basis of ANC. The phase inverted signal is added to wherever is being played back (or nothing if you are not listening to anything else). 

You're actually agreeing with me, whilst thinking you're disagreeing.

Noice cancelling headphones that also play a desired sound emit two waves simultaneously - principle of superposition and all that - one is the sound the person want's to hear and the other is a phase-inverted representation of the unwanted noise, ideally adjusted to compensate for the spectral characteristic of the receiver (microphone) and emitter (speaker).  The waves are added in software, and both are transmitted by the speaker.

It's exactly the same, except it's build in to the headphones and stuck on a persons head, and the speaker is also being sued to produce desired sound.

If the desired sound was silence, the headphone would produce an inverted version of the external noise, it is not filtering anything out. 

Edit: Re-wrote for brevity/clarity.

Is it not converted into (a very small amount of) heat?

Pedantic reply alert.  So easy to get things muddled.  Hope you don't mind...

Sound in air is always bleeding energy in to heat, but that's not behind areas of intensity in an antinoise system.  

Beyond that - primarily, no, the sound energy is moved away from the person's ear.  It may then be absorbed as heat however, but that can only happen because it's moved away from the person's ear in to a region of high absorption. 

It comes down to the geometry; considering a single frequency and a single emitter.  There's some volume everywhere in space, decreasing as you move further away from it.

Not split that source in to two physically separated emitters; there will be regions of space where their signals are out of phase and cancelling (sound is quieter) and other regions where the signals are in phase and reinforce (sound is louder).  

Moving to multiple sources produces an interference pattern in space, with the same total amount sound energy contracted into a smaller subset of that space, producing louder and quieter regions.  Think of a classing Young's double slit experiment (or google it) - uniform intensity light becomes a stripy pattern of brighter and darker regions.

Antinoise systems produce a quieter region in the person's ear and one or more louder regions somewhere not in their ear by displacing the sound energy through interference. 

Now, much of the displaced sound is presumably absorbed by all the dampening materials on the headphones, which then convert it to heat.

But the interference process itself doesn't convert sound to heat, it just displaces it.

Consider instead a region of two light waves interfering in a vacuums to produce an area of low intensity - there's nothing there to heat, so the energy can't possibly be turned to heat.

😵‍💫

Bet you're glad you asked the UKC experts first now, right! 😁 good knowledge as ever, Mr Wintertree. 

But it's done in software, of course it has to invert the signal, and it's added together in the phones themselves, it's not outputting an inverted signal to cancel the wave.

Open air systems use speakers to "play" the inverse noise to cancel it out.

I've completely failed to answer the important parts of the OPs question.  I should work on that...

Not true, what you are suggesting is the system monitored the external sound, inverts it transmits it out the speaker to cancel it out, adding what you want to listen to?

Why not, as many headsets do, monitor the external noise, invert it, add it in software, there by cancelling it out, then add in what you want to listen to, it makes no sense to add external noise and play that too, that's how you get more interference on what you are trying to listen to.

There isn't a massive need to transmit the inverted signal as you suggest, because the headphones themselves are like ear defenders, so the external noise is already attenuated, by the phones themselves.

Not if they are 180 out of phase there's no noisier or quieter regions, that's the whole idea.

Add sine and negative sine together = 0 the hard part is inverting the signal while keeping the phase time equal.

Exactly.

How does adding it "in software" go on to "cancel it out"?  It doens't.  The signal going to the speaker is a combination of the desired sound, and a inverted version of the external noise.  If you played that signal inside an anechoic chamber (or just a quiet room), you would hear the inverted noise.  

When the inverted waveform is played through the speaker, and equal and opposite sound wave is produced that physically interferes with the physical noise.

Not if you add noise that is of equal amplitude and opposite sign to the external noise.

What happens if you put noise cancelling headphones on and play silence instead of music?  You hear silence.  How?  Because the speaker in there is producing a sound wave that interferes with (cenacles out) the soundwave of the noise.

In some systems that will involve driving the speaker so it remains physically motionless instead of acting as a transducer passing external noise on to the ear.  For others, it will involve producing a waveform that cancels the noise at a location in space beyond the cone and closer to the ear.

Of course it does, adding two inverted signal then outputting the result gives you what?

Why would you measure the noise, invert it and then transmit it? You're performing and extra step, the transmitting, which isn't needed if you strip it out of what's coming out of the speakers first.

Why not remove it in software and only transmit, what you are interested in?

In the system your describing there's be no need to any attenuation of external sounds, but put a pair on without being plugged in or switched on, and they're as good as ear defenders

Are you talking about earphones / buds, rather than headphones?

Wrong.  Two separated emitters with some phase difference will interfere constructive at some points in space an destructively at others.  Because the phase difference at a given point is the sun of the phase difference between the sources and the phase shift due to the difference in the path lengths from each source to the point.

I think this is almost 100% an argument over terminology.  I repeat though: what signal is sent to the speakers if the desired sound played is silence instead of music?  

But no one mentioned phase difference, that's precisely why if you can do the inversion and addition in software, you don't need to worry about phase difference.

You effectively blot out all sound coming to the ear, you invert the ambient noise and add those two signals together, there by cancelling it out, then add in what you want to listen to.

Again, I'm talking about headsets here not earbuds.

I'm not making this up, I was involved in a system in the 1990's trying to quieten a compressor house

in a sewage treatment plant.

Nothing, there's not need. The sound is attenuated bu the headset itself, like simple ear defenders.

You only need to do something then there's something that isn't ambient noise.

I have also built such systems, but with a different kind of wave than sound.

I think this is largely a terminology issue.  I don’t doubt you know what you’re talking about, we’re just not communicating well I think…

Inversion *is* a phase difference.  It shifts all frequencies by 180 deg.

I agree. But the ambient noise is mechanical.  The DSP is electrical.  The noise has to be added mechanically to cancel.  The speaker has to be driven, hence my question of what signal goes to the speaker when the listener wants to hear silence.

As I said before, depending on the headset design that addition could happen at the speaker cone - meaning it’s driven equal and opposite to the external nose impinging on it to prevent it transducing external noise towards the ear, or it could happen at some plane beyond the speaker and towards the ear.

This was a post to Ciro illustrating how such a system displaces sound energy away from the ear rather than converting it directly to heat.  Assuming your microphone is some distance from the surface you want to nullify you’re still going to want to care about phase differences rather than do a trivial inversion, I’d kind of assumes modern devices have reference models for the various parts for this reason.

Nope. That's not true. I literally have a pair on my desk. With them on you can hear stuff. Turn ANC on (there is a switch on) and you can no longer hear stuff. That's with no sound playing.

I think what you are missing is that there are two sound sources.  The source of the external sound you want to cancel which is some distance away from the persons head and the sound from the speaker in the headphones.

The over-ear shroud of the headphones will significantly attenuate the external sound before it reaches the person's eardrum but some fraction of it is still present.  This is still undesired external sound, it is not coming from the loudspeaker in the headphones.

The desired music/speech from the headphones then has a phase-shifted and scaled version of the undesired external sound added to it.  The added signal is calculated to cancel the attenuated external sound *at the person's ear*.  So the person doesn't hear the undesired sound.

At some other point in space the external sound source and the 'cancellation' sound leaking from the headphones will not be 180 degrees out of phase and you could find a place where they may interfere constructively.  But in practice it won't matter because the noise cancelling waveform only has to cancel the attenuated external sound (after it passes through the physical damping in the headphone) so it is lower amplitude than the external sound.   Any of the cancelling waveform which leaks from the headphones will also get attenuated on the way out by the physical damping.  It is going to be so small relative to the external noise anywhere outside of the headphones it won't make much difference whether it constructively interferes at some point in space.

I have sony wh 1000xm2s and recently got the newer xm4s, and have used both with thicknesser, routers and circ saws. I won't say they are quite as good as some high attenuation ear defenders but certainly better than ear plugs.

The sound quality is superb, they've been what hifi's winner for a few years now

I've got the XM3s. Unless I've missed something, people are going on about NC producing silence, but with NC on, my Sony's broadcast a sort of quiet, swooshing white noise, very distinct when you first turn them on but you don't notice it after a while. 

What is this noise and why is it being made?

My best guess: the headphones aren’t perfect, and it’s harder to accurately cancel higher frequency noises because all the squishy real world factors become more important as the wavelength of the sound becomes smaller than the speaker coil, the microphone(s), the ear hole and so on; that’ll be frequencies over 3 KHz or so I imagine.

Interesting mixture of physics troll and actual request

https://isotunes.co.uk/products/isotunes-link-en352

Or you could cry a lot and get some Sony WH-1000-XM4s, which are apparently the dogs. Or look for bargains on the previous model XM3s.

I’m leaning towards some refurb xm4’s to give them a whirl.  30day exchange, so it might be worth a go.   The sound from the old 3M’s wasn’t great and I suspect if I try some Sony’s I’ll wonder how I managed😀

I do listen 6 or 7 hours a day, maybe I’ll treat myself. 

I wonder how using bone conducting headphones for music, with earplugs to block noise would work?

I've got a pair of cambridge audio melomania 2s which I find great as a pair of headphones for running or just around the house. Their passive noise reduction (from having something plugging my ears) is good enough for hoovering. However, they are a touch too big for putting ear defenders over the top for any length of time without getting uncomfortable, so for mowing the lawn I revert to my old system of bog standard ear defenders with a £20 pair of low profile wired earbuds. The attenuation of ear defenders + earbuds is pretty high, so I can listen to whatever I want at a sensible, relatively low volume. It's cheap, and if I run the cable up my shirt and out at the neck then I don't catch it on anything. It worked great building a boat with all the jigsaw work and hours and hours of powered sanding. I just needed some good five- hour podcasts to see me through! 

I think Simon and Garfunkel covered that one.

I recently got some XM4s. Maybe there is better out here, but to my ears the sound quality is fine. The noise cancellation is amazing. I have discovered that they are no good for Teams/Zoom meetings since I really can't hear myself talk! 

Turn the NC off😏