Why Does Sound Propagate?

[chconnor]

Let's all calm down a little

Nigel, and the others who aren't impressed with the thread:

While i'd make no claims as to the overall quality of the thread (my posts included), i think there is a very important exercise in it. Crashsite (and I, eventually, and maybe others) wanted to understand the propagation of sound on a particle-system level. Since all descriptions of sound we were taught or are able to find on the Internet involve abstracted models of air (the "masses on spring" model), we were at a loss and had to roll our own explanation. This, anyway, became the goal of thread, due to crashsite's apparent allergy to such explanations.

Unlike light, which occupies some nebulous wave/particle duality, air really is particles, and they have well-understood behavior, so it seemed a surmountable challenge. It may be a lot more functional to think of air as masses and springs, or pressure and waves, etc, but we sought a particle-level model that explained why the disturbance propagates at Mach 1, why it's independent of the initial disturbance speed, how the notion of "temperature" as kinetic energy played into it, etc. Even though we are obviously simplifying things a lot, we wanted to at least get together a rough and hopefully intuitive picture of the process.

Despite the low signal/noise, there's value in that.

A transducer displaces a medium, medium compresses, pressure waves move out at the speed of pressure waves in that medium. That's it.

That's what happens, no doubt. The question was "why". If it's obvious to you on a particle level why and how that happens, and why the system exhibits the counter-intuitive quirks that it does, kudos, and pardon us while we try to figure it out.

For most practical purposes the actual 'flow' of the molecules of air can be disregarded as inconsequential. [emphasis mine]

Well, it's not an exercise in practicality. It's an attempt to understand a process, not to use it build things.

Respectfully,
-c

 

[crashsite]

okay...my terminology sucks...

I do agree that the rarefaction is important to understand as well, but I'm not sure your explanation convinces me. To me, the propagation of the rarefaction is pretty easy to see (works on the same principle as the compression), but the way that the initial rarefaction imparts momentum is not as easy to see. I think of it this way: those bouncing balls are also trading momenta back and forth with the wall, or barrier (whatever is going to make the rarefaction) as they bounce off of it. When the disturbance happens, that movement imparts a negative momentum vector: the momentum of the balls bouncing off is now increased towards the disturbing barrier: with compression, the movement of the barrier is imparting positive momentum with every ricochet, with rarefaction, it is imparting negative momentum. The frenzy of balls in the next region out thus adopt that momentum towards the disturber and leave a similar void after them, and so on. Again, the intensity of the initial disturbance only determines how voided that area gets, and the overall temperature determines how quickly the next balls fill the void (and thus the speed of sound).

Yes, I do agree that I shouldn't have used the term, "sucks" with such a technically astute crowd. I should have left it that the inward excursion of the speaker cone leaves a void that the thermal activity of the molecules quickly fills...leaving it's own void which must be filled...leaving it's own void which must be filled (as the effect propagates away from the speaker).

But, this implies a "macro" action. In fact, it must be considered as an almost instant-by-instant action to really make sense.

(add on) Personal note to, chconnor. "He"...at least the last time I checked.

 

[Sceadwian]

Then stop the thread now because you need to go down to a quantum level if you want to understand the whole she-bang. It's pointless to discuss it in a casual manner like this and expect to get anywhere, you'll be bringing up different viewpoints and caveats from now till the end of the Universe and never get anywhere.

 

[crashsite]

ad infinitums

Then stop the thread now because you need to go down to a quantum level if you want to understand the whole she-bang. It's pointless to discuss it in a casual manner like this and expect to get anywhere, you'll be bringing up different viewpoints and caveats from now till the end of the Universe and never get anywhere.

I feel like we're getting somewhere. I also feel like there is an endpoint (at this level of the discussion). As to whether it will lead to a follow-on discussion at an even more elementary level, I can't say but, so far, I haven't really felt the urge for it.

Oh, stop cheering already, guys....

 

[chconnor]

Then stop the thread now because you need to go down to a quantum level if you want to understand the whole she-bang. It's pointless to discuss it in a casual manner like this and expect to get anywhere, you'll be bringing up different viewpoints and caveats from now till the end of the Universe and never get anywhere.

A fair argument. Akin to 1st-graders having a somber conversation about politics or something: they don't realize how ridiculous their models are. I can understand all the non-helpful posts people submitted that essentially said "don't worry about it" or "don't think about it on that level", if that in fact is their perspective of us.

For the record, the idea was never to comprehend the whole shebang at all, it was just to get to the particle level with a simplified model that helps us understand the basic phenomena "beneath" the level of mass/spring (at least, that was my goal.) I'm satisfied for now, and it sounds like crashsite is too, even though there undoubtedly loose ends all over the place.

Maybe we're terribly misguided, but if a simple mass/spring analogy is a fair-enough approximation of the situation, complete with simplified/rough equations to represent/estimate speed of propagation, energy dispersion, temperature, etc, then it doesn't seem far-fetched that a particle-level explanation might have a similarly-simple analogy, even if it's not quite as simple (this was crashsite's hunch earlier in the thread.) I dig the mass/spring analogy, but it was an inviting challenge to try to understand it on the particle level.

If, in fact, crashsite and I are 1st-graders theorizing about politics, then please, smarter people of the Internet, either provide a particle-level analogy that makes sense, tell us how our respective models are flawed, or provide some kind of convincing reason (ideally, with examples) for why the situation is irreducibly complex and mas/spring is our only hope. All crashsite was asking for was something besides dismissive hand-waving and rehashed one-line textbook explanations of the phenomenon; i respect that, even if he and i weren't always on the same... wavelength.

And seems to me that even the depth of the quantum perspective isn't the "whole shebang" either, but that' s different thread.

-c

 

[crashsite]

Okay, I admit it...If I were a US diplomat, we'd already be in WWIV

A fair argument. Akin to 1st-graders having a somber conversation about politics or something: they don't realize how ridiculous their models are. I can understand all the non-helpful posts people submitted that essentially said "don't worry about it" or "don't think about it on that level", if that in fact is their perspective of us.

For the record, the idea was never to comprehend the whole shebang at all, it was just to get to the particle level with a simplified model that helps us understand the basic phenomena "beneath" the level of mass/spring (at least, that was my goal.) I'm satisfied for now, and it sounds like crashsite is too, even though there undoubtedly loose ends all over the place.

I must say that you certainly are a lot more tactful than myself. Must be in the genes. But, I try to make up for my lack of tact with sheer cantakerousness...which is often misinterpreted as being...well..."sheer cantanderousness".

I suppose one could view the interaction of the particles as being elastic (ie: they don't actually strike each other, rather they get close enough that their repulsive force pushes them apart again). I saw an old PBS science show where this was somewhat comically portrayed by a baseball announcer telling about how the bat only approaches the ball. But, whether the particles collide or merely almost collide is really moot in the context of explaining sound propagation. Likewise, how the energy from the surrounding environment actually gets into the particles to make them vibrate is moot. I mean, it's important, just not to the issue of how the sound propagates.

Okay, let me stick my foot in it again. I'm sure I'm going to get a few choice comments here but, it's looking like sound is propagated by heat. The vibrations of the molecules by thermal energy pushing the sonic wave front through the air seems to be almost as mechanical as the combustion of gasoline pushing on the pistons of a car engine ends up pushing the car down the road.

I apologize to all you "high-level thinkers" but, I'm the kind of guy that really does need to beat these things down to their very lowest common denominator in order to feel like I understand them (note emphasis).

Let me again reiterate this point. All the "spring and mass" stuff doesn't really explain any of sound propagation. It does a good job of explaining how waves work but there's just no waves involved in sound propagation. I'm not mathematically inclined enough to say for sure but, I stronly suspect that it only works becuase of the random nature of the vibration of the molecules and that any embedded wave structure would interfere with the process.

Okay, let me jump in here before you do on the issue of, "waves". Yes, there are waves associated with sound and there are standing waves present in the air when a sound is present and we can use a microphone to pick up and record sound waves. But, the waves are an artifact of the spacial time/distance continuum of the phenomena of sound, not part of the actual process of sound propagation.

I don't understand why everybody here seems to want to think about this on the macro, spring and mass level and why nobody seems to want to think about sound propagation on an instant-by-instant basis when it seems like that's where the real action is.

 

[3v0]

But, the waves are an artifact of the spacial time/distance continuum of the phenomena of sound, not part of the actual process of sound propagation.

You can not dismiss the physics that makes waves... waves.

The underlying mobo-jumobo that stores the energy is important. But it is the physics of wave montion, the conversion between potential and kenetic energy that propagates the sound.

 

[chconnor]

Okay, let me stick my foot in it again. I'm sure I'm going to get a few choice comments here but, it's looking like sound is propagated by heat.

It occurred to me as well that sound wouldn't propagate without heat. As in, absolute zero temperature, totally-unmoving particles, thus no transfer of momentum except from the diffusive effects of the immediate momentum-trading inital collisions, which quickly loses its locality and directionality. I think "sound is propagated by heat" is a great way to put it.

I don't understand why everybody here seems to want to think about this on the macro, spring and mass level and why nobody seems to want to think about sound propagation on an instant-by-instant basis when it seems like that's where the real action is.

I think it's just a question of what you need to be satisfied that your model is deep enough. The billiard ball/particle system certainly isn't the bottom of the story, it's just another rung down the ladder.

Just for one example: PBS was right, as far as my limited understanding of particle physics is concerned: air molecules never actually collide, we just describe it that way because we can simplify the complex situation (yet another thread) by saying that they "collide" and "bounce off each other". It's a reduction of detail to suit the purpose. Another, higher-level reduction, is to group particles together and say that these metaparticles have mass and springs, etc etc. So it's arbitrary where on the ladder you want to build your toy model, as long as the model satisfies whatever needs you have for understanding/computation/whatever.

Anyway, I know that's no surprise to anyone. Generally, I think most people don't really care how the wave propagates on the molecular level (as many have made clear in this thread) because it doesn't seem "useful" to worry about. I was curious, personally.

-c

 

[chconnor]

It occurred to me as well that sound wouldn't propagate without heat.

Check out http://www.sengpielaudio.com/calculator-speedsound.htm

...if it is to be trusted, you can enter .1K, .01K (~2m/s!), or 0K, and watch the speed of sound go to zero... no surprise to physicists, perhaps, but i was pleased to see that.

-c

 

[chconnor]

You can not dismiss the physics that makes waves... waves.

The underlying mobo-jumobo that stores the energy is important. But it is the physics of wave montion, the conversion between potential and kenetic energy that propagates the sound.

I agree with both 3v0 and crashsite, because i think it's just semantics and models working at different levels in the same universe. Or, have i said that before.

-c

 

[3v0]

I agree with both 3v0 and crashsite, because i think it's just semantics and models working at different levels in the same universe. Or, have i said that before.

-c

You can't agree with both of us in any sane way. Crashsite said waves are but an artifact, I say they are the means by which waves propagate.

 

[chconnor]

You can't agree with both of us in any sane way. Crashsite said waves are but an artifact, I say they are the means by which waves propagate.

Let me refine what i meant. I agree with both models (more or less; crashsite and i aren't on exactly the same page) meaning: I agree with crashsite that "particle interaction" causes propagation, and I agree with 3v0 that waves cause propagation. I don't agree that the wave perspective is the only valid mechanism for understanding the propagation, or that particles are a "more real" perspective for the propagation (not trying to put words in anyone's mouth here).

-c

 

[3v0]

It is the physics of wave montion, the conversion between potential and kenetic energy that propagates the sound.

The "particle interaction" is just the device that allows the energy conversion with sound waves.

 

[crashsite]

Auld Lang Syne

...it is the physics of wave montion, the conversion between potential and kenetic energy that propagates the sound.

Is it?

It's easy to see how waves exchange potential and kinetic energy to replicate and propagate more waves but, it doesn't seem to be a good explanation for the propagation of sound (at least not the sonic wave front).

I do agree with you that we can't both be right. I was hoping that my "pool table" post would show that my view is the correct one.

I can see you're not impressed. That's okay. I know this exercise has improved my understnding of things immensly (even if it has left me still pig-ignorant), and I do appreciate your input and views as well as all others. It really is all grist for the mill and even in disagreement, provokes thought.

I agree with, chconnor that this thread has probably run its course but, with an awful lot of loose ends (perhaps a lot of awful loose ends) so, I'm sure we'll be on opposite sides of other discussions in the future here.

 

[crashsite]

Sound Propagation - Tying Up Loose Ends - Thermal Effects

I have been continuing to mull some of the implications of the, “sound propagation” thing and want to take a stab at tying up some of the loose ends. For Those here who find such efforts useless or such a post, “pointless” and “pure crap”, you may as well stop reading so as not to waste any more of your valuable (or even valueless) time. But, thanks for stopping to look….

This post is my take on things and deals only with some of the, thermal effects. Here's the link to the reference post (near the bottom of page 14 and titled, "Got It"):

https://www.electro-tech-online.com/threads/why-does-sound-propagate.87198/

The following few posts in that thread clarify some things.

---------------------------------------------------------------------------

A comment was made that, during the compression cycle, the sonic shock wave will tend to overtake itself and during the rarefaction cycle, will spread out. Actually, I had already thought of that. In thinking about the sonic wave front being propelled by heat, that really does make sense. As the air compresses it heats and the Mach speed increases. As the air rarefies, it cools and the Mach speed decreases.

But, that analysis brings up another logical conclusion. This may very well be a simplified description of adiabatic heating and cooling. In aviation we learn that the atmosphere normally cools with altitude (setting aside specific conditions that allow temperature inversions). It seems reasonable that simply having more or fewer air molecules bouncing around is the same as having more or less heat.

In nature, sounds are generated and they propagate out into the environment where they may be heard or otherwise affect things. Those sounds may be generated by natural or man-made sources; the chirp of a cricket, the babbling of a brook, a peal of thunder, the wistful whistle of a distant steam locomotive…and so on. But, whatever the source, any environmental factors that affect the propagation are also natural and part of the process. Essentially, regardless of the sound or how it propagates, it’s 100% fidelity.

But, if a fastidious audiophile puts a microphone into that environment and records the sounds and then plays it back later, there’s an inherent distortion. A distortion that I’ve never heard of anyone ever mentioning or trying to correct. As noted above, during the compression cycle of the sound, the air heats slightly and the sound propagates faster and during the rarefaction cycle, slower. The microphone faithfully records the sound, with that bias.

When the audiophile plays back the sounds, the compression and rarefaction Mach speed effects are essentially doubled…and, that’s only if the sounds are played back at the original decibel level. If the playback is louder, that distortion increases even further. How can a truly fastidious audiophile live with such a breach of fidelity? He must surely design a recording/playback system by which the original sound pressure levels are recorded (on a cycle-by-cycle basis) along with the audio content and then design a digital compensation circuit for his pre-amp to correct those phasing and time-shifting distortions. A Simple filter wouldn‘t do since the effect occurs with each and every cycle of the audio signal!.

BTW: I’ve dubbed that distortion the, Sala Effect (named after myself) and note it here to put a “time stamp” on it.

(add on) Coming back to aviation, I'm beginning to wonder if heat is not a pretty good explanation of how an airplane flies. It's an intriguing thought (intriguing to me, anyway). If it turns out to be true that heat is at least partially due to the density of the molecules (in addition to their level of activity) then an airplane wing may use this effect to fly. Certainly, an airplane in flight is not materially heating up the atmosphere so, we need to think of the local effect.

By separating the airflow, with some going over and some going under the wing, the two airstreams can be examinted independently. By forcing the air over the top surface into a longer path (the usual, simplified explanation), the air molecules spread out and thus have a lower pressure. But, why is there a lower pressure? The molecules still have essentially the same amount of thermal energy and that results in impacts pushing downward on the wing. But, since there are more molecules on the bottom side of the wing, the sum of the combined force of their thermal impacts, pushing the wing up, is greater than the sum of the thermal impacts above the wing pushing it down. Ergo, simple Newtonian action at work.

I admit that there is a, "density altitude" issue that's unresolved in the above analysis and I do have some thoughts about it but, I think I'll leave it as an 'action item' to ponder for now as I would be interested in other peoples' notions about it...assuming there are some.

 

[Sceadwian]

Electret mics can record absolute pressure values.

 

[crashsite]

Electret Microphone

Electret mics can record absolute pressure values.

But, can it tell a playback processor what those values are?

The electret microphone is just a variation on the condenser microphone. Same operation but, instead of applying a voltage to the pickup element, the voltage is "stressed into" the dielectric during manufacture so no voltage is needed. The low voltage the electret microphone uses is used to operate a FET that's needed as an impedance converter to minimize hum pick up.

I'm not sure why an electret microphone would be any more or less capable of sensing absolute pressure changes than any other microphone but, I'd be very interested to find out.

 

[Sceadwian]

You'd have to avoid using any decoupling capacitors on the input of the amplifier(s), but the electret microphone will produce a DC bias based on absolute pressure. The capsules themselves are hermetically sealed to keep in that static electric bias.

 

[crashsite]

DC vs. AC

You'd have to avoid using any decoupling capacitors on the input of the amplifier(s), but the electret microphone will produce a DC bias based on absolute pressure. The capsules themselves are hermetically sealed to keep in that static electric bias.

I'm not sure that's true. The first point is that the Sala Effect© varies with frequency. Well, not really but, on a percentage basis it does. If a 200 Hz wave is displaced by a few microseconds, the effect is nil but if a 12,000 Hz wave is displaced by the same amount the effect is 60X greater. That means that only AC levels need be addressed and capacitive coupling would work. The electronics for integrating the effect into the recorded audio, when using digital techniques, would not be difficult. A tremendous amount of stuff is already embedded in both audio and video disks along with the program content already.

But, the nuts and bolts of "how to" is really a separate matter and not related to the physics of the effect and need to null it for 'proper' audio fidelity.

I'm also not completely convinced that an electret microphone is able to sense and transmit absolute pressure levels. I know that, years ago, back when FETs were coming online, there was talk of having such high impedances that true electrostatic charges could be sensed but, I don't know what the status of that is even with MOS-FETs as the impedance converter.

But, the idea there was to sense the change in dielectric constant of having more or less air (usually air was the expected medium) between static plates rather than having the plates move relative to each other. Since the electret has a physical movement of at least one surface, and there some elasticity, that's why I question if absolute pressure can be detected that way.

I had thought to use that dielectric only feature in an aircraft altimeter that would have no bellows or other moving parts at one time...

 

[Sceadwian]

**broken link removed**

How does this effect fidelity? Easy answer, it doesn't. The shape of the human ear itself has more effect on the perception of sound than the frequency dependent component. Last time I checked for true in your face fidelity the microphone has to actually be placed inside the ear canal, or in mock ear canal very similar to the individual and played back inside the ear canal to notice these effects, and they only alter spatial perception of audio.