Anyone Into Noise Cancelling (In a BIG way)

[Old Raft]

Yesterday I tried to make a phone call in an office next to people bashing 1" dia rivets to death.......... it's kind of loud (the riveting not my phone call)  the sort of thing that would make a 747 passing overhead at 100 feet seem quiet.

Is there a way to cancel the sound by just inverting the received noise and transmitting it back via a large speaker? 

[Hugo Mills]

Sudden percussive sounds with lots of randomness in... no.

Low-level (and low-frequency) continual hum with a strong repeating
component (at the waveform level), yes -- but you can't use a
loudspeaker to do it, because you'll end up with interference and
phase variations where it makes it quieter in some places and louder
in others. To make this work, you need headphones very close to the
ear and some serious DSP work. This is what noise-cancelling
headphones do. Excellent for things like being inside an airplane,
much less useful for being inside a riveting shop.

Hugo.

--
Hugo Mills | Biphocles: Plato's optician
hugo@... carfax.org.uk |
http://carfax.org.uk/ |
PGP: E2AB1DE4 |

[Eric Rowen]

It had occurred to me that the noise cancelling would try and cancel itself as it were :-)  or certainly cause a lot of phase difference problems. 

--
You received this message because you are subscribed to the Google Groups "Reading Hackspace" group.
To unsubscribe from this group and stop receiving emails from it, send an email to reading-hacksp...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Stuart Livings]

Hi,

I'm afraid that you're not going to be able to solve the problem this way for a couple of reasons.

1) Noise cancelling works by generating a noise 180 degrees out of phase with the undesireable noise to intersect perfectly at a given point.  If the listener is in a different place to that point then the effect is lessened and may even amplify the undesireable noise.  The wavelength of the sounds gives you an idea of how accurate the intersection needs to be.  At 1kHz the wavelength is approximately (330mps / 1,000 = 0.3m) 30cm, so if you move 15cm you'll actually double the noise!  Sharp sounds have a very high frequency, meaning that this distance gets much shorter.

2) Because you have to generate the complementary sound a noise cancelling system has to "hear" the undesireable sound and process it to generate the desired complementary sound.  For sharp sounds the processing system has an impossible task: To identify and generate the complementary/out of phase sound while the undesireable sound is still propagating to your ear.  Noise cancelling headphones don't normally prevent a listener hearing speech or other ambient noise for this reason.  Extremely sharp sounds will often cause 'clicks' to be heard by the listener as the cancelling system generates the out-of-phase sound slightly later than the undesireable sound.

The good news is that while you cannot solve your problem with noise cancelling systems it is a solved problem using other (simpler) technology.

Disposable foam earplugs are extremely cheap and, because they're a passive barrier to the sound wave, work very well for sharp sounds.  SImilarly ear defenders work very well if you have an aversion to sticking something in your ear.

Both of these can be found at many industrial tool vendors, including online.  I personally like the LaserLite foam earplugs, I find them very comfortable and wouldn't hesitate to recommend them.

Stuart

On 22/06/2016 03:05, Old Raft wrote:

Yesterday I tried to make a phone call in an office next to people bashing 1" dia rivets to death.......... it's kind of loud (the riveting not my phone call)  the sort of thing that would make a 747 passing overhead at 100 feet seem quiet.

Is there a way to cancel the sound by just inverting the received noise and transmitting it back via a large speaker? 

[Hugo Mills]

On Wed, Jun 22, 2016 at 03:28:56AM -0700, 'Stuart Livings' via Reading Hackspace wrote:
> Hi,
>
> I'm afraid that you're not going to be able to solve the problem
> this way for a couple of reasons.
>
> 1) Noise cancelling works by generating a noise 180 degrees out of
> phase with the undesireable noise to intersect perfectly at a given
> point. If the listener is in a different place to that point then
> the effect is lessened and may even amplify the undesireable noise.
> The wavelength of the sounds gives you an idea of how accurate the
> intersection needs to be. At 1kHz the wavelength is approximately
> (330mps / 1,000 = 0.3m) 30cm, so if you move 15cm you'll actually
> double the noise! Sharp sounds have a very high frequency, meaning
> that this distance gets much shorter.
>
> 2) Because you have to generate the complementary sound a noise
> cancelling system has to "hear" the undesireable sound and process
> it to generate the desired complementary sound. For sharp sounds
> the processing system has an impossible task: To identify and
> generate the complementary/out of phase sound while the undesireable
> sound is still propagating to your ear. Noise cancelling headphones
> don't normally prevent a listener hearing speech or other ambient
> noise for this reason. Extremely sharp sounds will often cause
> 'clicks' to be heard by the listener as the cancelling system
> generates the out-of-phase sound slightly later than the
> undesireable sound.

Even at low frequencies, there are some sounds that the noise
cancelling can't cope with.

There's a fascinating effect with my noise-cancelling 'phones (AGK
470, IIRC), where if you're sitting on the back of a bus and it goes
over a large pot-hole, you get a throbbing "wub-wub" noise from the
noise cancelling. I think it's due to the frequency that the windows
wobble at. It's completely undetectable without the 'phones, though.

They also produce a ~1.5 - 2 Hz throb on small sprinter trains when
the windows are open. Again, largely unnoticeable without the noise
cancelling, but very pronounced with.

> The good news is that while you cannot solve your problem with noise
> cancelling systems it is a solved problem using other (simpler)
> technology.
>
> Disposable foam earplugs are extremely cheap and, because they're a
> passive barrier to the sound wave, work very well for sharp sounds.
> SImilarly ear defenders work very well if you have an aversion to
> sticking something in your ear.
>
> Both of these can be found at many industrial tool vendors,
> including online. I personally like the LaserLite foam earplugs, I
> find them very comfortable and wouldn't hesitate to recommend them.

They do make it difficult to have the phone call, though. :)

Hugo.

> Stuart
>
>
> On 22/06/2016 03:05, Old Raft wrote:
> >Yesterday I tried to make a phone call in an office next to people
> >bashing 1" dia rivets to death.......... it's kind of loud (the
> >riveting not my phone call) the sort of thing that would make a
> >747 passing overhead at 100 feet seem quiet.
> >
> >Is there a way to cancel the sound by just inverting the received
> >noise and transmitting it back via a large speaker?
> >
> >

--

[Matthew Daubney]

Buy a phonebox for the office and insulate it from sound!

Also, dress it up like a tardis. JDT

[Eric Rowen]

But assuming I did want to try (sound levels and source vary a lot) where would I start? 

[Ryan .]

What Matt said.

Anything else is a huge deal. (Where I would start is a model of the system, from that you can get the specs you'll need to implement in hardware)

[Eric Rowen]

Earplugs and ear defenders are pretty much  de rigueur and as I have other interests involving loud noises then I have found that the smooth foam EP-509  (  too soft and spring back into shape before they get into your ear ) are inferior to  "Classic E*A*R  Aeraro which slowly expand after rolling between finger and thumb and fit and seal very well, the sound goes as you feel them expand. 

Unfortunately ear defenders and plugs kind of negate a phone call, unless there is an earplug incorporating a telephone earpiece?  

The  

"They also produce a ~1.5 - 2 Hz throb on small sprinter trains when

the windows are open. Again, largely unnoticeable without the noise
cancelling, but very pronounced with.

is interesting as this is well below the threshold of hearing and sounds like a beat produced by the interference between two already low frequency sounds, as suggested like the windows vibrating and some other train related low frequency sound. 

[Eric Rowen]

Lash ups are one thing, but I have never been into "modeling" systems so I would need some pointers here. 

[Ryan .]

Bose/other noise cancelling headphones are great and work well if you want something active, or you could use passive noise cancelling techniques (think double-thickness plasterboard) to make a quiet space. 

As for doing it on a wider area in an active way, see other responses for why that won't work. 

As for modelling systems generally...Hugo? :) Scilab and similar are your friends, https://en.wikipedia.org/wiki/Mathematical_model

[Stuart Livings]

The LaserLite plugs I use are absolutely fine for phone calls.  Voices are a little muffled but their filter frequencies don't affect voice that much.  I use them on aircraft and can just about hold a conversation with the stewards even despite the background noise if I can get the visual cues of lip-reading.

Next time I'm in RLab I'll drop a few sets in for you to experiment with.

Stuart

[Eric Rowen]

More thought on "sudden percussive sounds"  these are bursts of regular volume noise, but probably start with a very sharp rise in sound, a little like a square wave  so (in this simplified approach) containing all the odd harmonics at higher frequencies than the implied sine wave.  I'm not sure this helps !!  :-))   but it adds another dimension, rather than a simple single frequency and 180o out of phase approach. 

[Hugo Mills]

On Wed, Jun 22, 2016 at 01:35:44PM +0100, Ryan wrote:
> Bose/other noise cancelling headphones are great and work well if you want
> something active, or you could use passive noise cancelling techniques
> (think double-thickness plasterboard) to make a quiet space.
>
> As for doing it on a wider area in an active way, see other responses for
> why that won't work.
>
> As for modelling systems generally...Hugo? :) Scilab and similar are your
> friends, https://en.wikipedia.org/wiki/Mathematical_model

Personally, I'd probably start with Mathematica (free, if very
slow, on the RasPi), although NumPy+SciPy or MatLab would also be
options. I haven't done a lot of that kind of modelling, though -- I'm
much more discrete than continuous. :)

I feel that the optimal solution to this problem is simply to put
as much space between the source of the sound and your ears, or to
stop the source entirely. I don't think there's likely to be a good
technical solution which will solve the issue. Is this the result of
temporary works, or is it a permanent part of the environment? (And in
either case, if the noise is so loud, what safety measures have the
owners of the facility put in place to mitigate the prospective
hearing damage of people in that office?)

Hugo.

Hugo Mills | "I will not be pushed, filed, stamped, indexed,
hugo@... carfax.org.uk | briefed, debriefed or numbered.
http://carfax.org.uk/ | My life is my own."
PGP: E2AB1DE4 | Number 6, The Prisoner

[Eric Rowen]

Permanent source, either removing rivets, replacing rivets on peening over the heads ............. isolation from isn't a possibility, protection both earplugs and ear defenders are provided and part of H&S equipment, but other than sign language at close quarters or Email not phone then the noise is an everyday feature. 

[Chris M]

In-ear headphones, ear defenders and a throat mic?

[Matthew Daubney]

Or http://www.onedirect.co.uk/peltor/3m-peltor-flex-ear-defender-headset?gclid=CLbe_7eUvs0CFYcp0wodUhQElA&LGWCODE=PELTORHF%3b55331%3b2389

No idea how well they work though. Also £xxx

[Norro]

Hearing aids work with a tube through the earplug it seems to me that a speaker for a phone could be piped through a deferder in a similar way without lessening the protection

[Eric Rowen]

I like that idea..... worth investigating ! 

--

[Robert Pryde]

How about ear plugs designed for shooting ranges : http://www.ultimateear.com/products/reactear.php

They let normal level noise through but blocks gunshots.

On Wednesday, 22 June 2016 11:05:46 UTC+1, Old Raft wrote:

[Eric Rowen]

mmmm... I'm so used to the ear defender (external) earmuff  types that I had not thought of the internal version.

My one caveat would be beware of the "fitting" procedure, they insert a plug into your ear to stop the moulding compound getting into your inner ear before it sets, however the plug (cotton wool or whatever) isn't secured to anything and if they insert it too far it disappears inside your ear (yes it did ! :-)   This is uncomfortable and dangerous.  

[Ryan .]

I made my own with a £10 kit from eBay, worked a treat for the bike but they don't have the fancy electronics.