Why Does Sound Propagate?

[crashsite]

It's just not as simple as a wave...

When you compress a gas it heats up. simple law of physics.

That's why I was careful to bring the air back to a specified temperature after compressing it. I assume there to be some sort of Mach number change during the compressing operation that tends to mask the pressure vs. thermal effect.

Sound is also a series of waves that travel by a compressive action of sorts.
So yes in a way you are right about the thermal interaction of sound with air. Still a known principal though.

When you say that "Sound is also a series of waves...", you have to be careful not to get into semantics. Yes, "sound" is waves that have acoustical properties that are heard as sound. But, a sonic wave front isn't necessarily "sound" by that definition. It can be a pulse dragged along by a supersonic wing, or ultrasonic energy produced by a leaking gas line or a click from the little Digitape electronic tape measure I showed in another thread, etc.

But, regardless of whether it's "sound" or some other sonic phenomena, it still propagates the same way through whatever medium.

When thinking of a wave, half the time it's rarefaction. In fact, over time, unless there's a net increase or decrease in pressure, exactly half the time. But, rarefaction is one of those things that brings up more questions. Certainly, there's an adibiatic cooling that occurs, at least locally near the "speaker cone" (keeping "air" and "speakers" in play), as the air is pulled to a partial vacuum. But, even pulling a vacuum takes power and that power needs to be accounted for.

For a real life demonstration of thermal acoustic effects go to a car audio crank it up contest.
Do a temperature measurement of the air in one of the highest powered vehicles you can find. Then do another test just at the end of a run. You will find a several degree temperature rise just from the intense compressing and vibrating or the air in the vehicle. More so than what thermal radiation from the speakers and amplifiers would account for in that time period.

I've never done that but, I'm not sure how you'd end up with more energy (heat) than is generated. Maybe the power transister heat dissipation does add some heat. Resistive dissipation of the voice coils and wires would add some. Friction effects such as might be within the foam surround of the woofer cones, that doesn't add much to the audio output could account for some heat, etc.

Actually, I'm a proponent of limiting the power of car stereo systems. I propose an ordinance that allows a person one RMS Watt per IQ point. I figure that would quiet most of the systems considerably.

Just an observation to help you out. I dont think your wrong on your question at all. I do however think you may just be looking at the wrong places for your information.

I think I make it pretty clear that I'm at least suspicious of the info given in the Wikipedia on the subject and, since that seems to be the prevailing notions of how it works, in the world of physics...pretty much all other sources, too.

So basically, I'm not just sure where else I should be looking...

Perhaps using 'Thermal acoustic effects' as a way to narrow your search will help.

For a possibly better or more application specific device that uses thermal acoustic effects to do work read up on hot air engines. There are thermal acoustic resonance engines that use nothing but heat and a tuned chamber to drive a piston and produce mechanical motion. They are a cousin to the sterling engine. They are very simple and easy to build too!
And being a centurys old design there is a good deal of documentation and written work on how they do what they do with heat and sound.

But, if the question is, sound propagation would I not be likely to complicate my thinking considerably by trying to reconcile the operation of a heat engine with the generation and launching of a sonic wave front? Even accounting for the "tuned" intake and exhaust characteristics that are taken advantage of in many different types of engines (2 stroke and pulse jet, notably). In my original thread on this topic I did comment, in one post, that I had concluded that sound is propagated by heat just as a car is propelled by the heat of combustion pushing on the pistons in a car engine.

 

[crashsite]

Sound Propagation - A Question of Vacuum

Yeah, back to this again. I must say that I've been somewhat disappointed in the sound propagation "answers" so far and, I'm not sure this is the correct venue for this question either but, I'll try it here.

A common physic lab project is to demonstrate that sound wont travel through a vacuum by pumping down a bell jar with t sound source in it.

The question is, if anyone here knows at what level of vacuum the sound will no longer propagate through the air and the characteristics of how the propagation of sound falls off.

Does it become weaker as the pressure is reduced? Does it remain pretty constant and then fall off with some slope at some level of vacuum? Does it remain pretty constant until the molecules get too far apart to sustain propagation and then sharply fall off?

Also, what are the characteristics of air near the point at which it will no longer support sound propagation?

 

[dknguyen]

This sounds just like one of Jasonbe's questions.

I'm not even sure why you are asking this since you know that sound NEEDS air to propogate. Period. I know that you know. From that falls some very obvious conclusions:

-A vacuum chamber with a membrane on one of the walls vibrates- no sound generated or propogated. Period.
-Now if you had an invisible forcefield that produced a clearn line between air and vacuum, the sound would come to a dead stop with when it hit the vacuum. No air, no sound. It wouldn't decay. It would just stop like if you ran a car into the wall.
-If you had a "diffusive forcefield" produced a pressure gradient that approached a vacuum, the sound would get weaker as the air got less dense.

As for the "level of vacuum" it's probably a decaying exponential. So the "threshold" you want depends on how you define the threshold. Two air molecules in what is otherwise a vacuum will technically "conduct" sound but will do so at a fantastically low probability, over an almost null distance, with a ridiculous amount of decay.

Also, what are the characteristics of air near the point at which it will no longer support sound propagation?

*cough* jasonbe *cough* Less air is less air. THere's no big mystery about it. What are the characteristics of copper near the point at which it will no longer support conduction?

 

[Hero999]

As the air is removed I imagine that higher frequencies will be affected more than lower frequencies. This is because the spacing between the air molecules gets larger compared to the wavelength.

 

[dknguyen]

As the air is removed I imagine that higher frequencies will be affected more than lower frequencies. This is because the spacing between the air molecules gets larger compared to the wavelength.

The spacing? Or harmonic oscillation?

 

[crashsite]

As the air is removed I imagine that higher frequencies will be affected more than lower frequencies. This is because the spacing between the air molecules gets larger compared to the wavelength.

Okay, I want to take your response first since it gets right to the nitty gritty of it.

As you've probably figured out by now, I do not agree (and the more I learn the less I agree) with the way sound is thought of and taught by the physics community. Your answer provides an excellent opportunity to delve into what may actually be happening.

You're squarely back at, "wave analysis" when I believe the issue is, "sound propagation". Having said that, I'm going to stop and let you reconsider your thoughts. If you still believe what you said, after thinking aobut it, then I'd like to discuss the details of why you think wavelength has anything to do with how the sound propagates.

This is at the very crux of my question on this subject of, "why does sound propagate".

 

[Hero999]

I was wrong there.

As the air pressure drops, the speed of sound will also be reduced meaning the waves will get longer.

My gut feeling is still telling me that higher frequencies will be attenuated more but I can't explain why. My inner voice is failing me.

 

[dknguyen]

Intuitively, I think higher frequencies just dissipate more quickly because there are more collisions going on so the more energy is dissipated over a shorter distance. Can't say for sure though. It also happens with mechanical vibration, light, and water waves so there's probably something fundamentally common there.

 

[Hero999]

If it's dissipated more quickly doesn't that mean that the energy is also lost more quickly?

Why is it that dense substances can pass very high frequency sounds but air can't?

A 5MHz sound will travel a reasonable distance in water but will hardly travel through air at all.

 

[dknguyen]

If it's dissipated more quickly doesn't that mean that the energy is also lost more quickly?

Yes.

Why is it that dense substances can pass very high frequency sounds but air can't?

Less damping? Because of tighter molecular bonds?

A 5MHz sound will travel a reasonable distance in water but will hardly travel through air at all.

I think it's wrong to put it like that, because sound of any frequency will travel farther in water than air.

I've been reading up on MAVs and the REynold's number and it seems that might come into play here somewhat...The ratio of the inertial forces of a fluid vs vs the viscous forces. Inertial forces (that keep the fluid moving) are proportionally greater than the viscous forces (that stop the fluid from moving) when the fluid is denser, when the object moving through the fluid is larger or faster. Basically that means more air, once moved, stays moving for a longer distance.

 

[Hero999]

But water has a higher 3dB roll-off than air.

I reckon that 3dB roll-off gets higher as the air pressure increases because the conductivity increases.

 

[dknguyen]

But water has a higher 3dB roll-off than air.

What are you saying "but" to?

 

[crashsite]

Apples and Oranges

The spacing? Or harmonic oscillation?

First, let me reiterate that this is not a "Jasonbe issue". His game is to garner attention by dodging his own questions in ways that drag people into trying other approaches for helping him. My intent is to understand sound propagation on a conceptual (rather than mathematical) level. The reason it can look like Jasonbe is because there is such a strong bias for "wave analysis" when the issue is "sound propagation" and I'm not willing to accept some crappy answer about waves, schmoozed over with mathematical machinations, to gloss over what's really happening.

I believe it's obvious that, if sound is propagated by molecular collesions, that you were right regarding the "falling off" of sound intensity as a function of the probability of molecules colliding. I too, do not know what that curve would look like but, I don't need to know that bit of data to feel like I have an understanding of the concept of it.

But, while there may be less sound because of fewer molecules, the speed and nature of the propagation remains the same. Even when the molecules are far apart, if they can collide at all, the sound manages to still propagate at the same speed. I'm still convinced that the issue is not "balls separated by springs" and the interchange of potential and kinetic energy that's at work there.

Let me mitigate that last statement somewhat by stating that I do accept that, since molecules don't actually collide but rather experience a repulsive force as they near each other, that there is some ball and spring effect going on at some level. I just don't know if it's directly associated with sound propagation or not.

That you even suggest, "harmonic oscillation" troubles me that, in your mind, wave analysis is the dominating thought.

 

[dknguyen]

So how would you look at something that's moving back and forth? Or what would you call it? Because that's what's happening. A lower frequency means the air molecule is moving back and forth over a greater distance. Nothing schmoozey or mathematical about that. Or I could just replace all those words with "harmonic oscillation".

Ask yourself, would you have thought about it any differently if I had say it the second way as opposed to the first way? Even though they mean the same thing? If it does, something's wrong with how you're thinking about things.

I'm not actually sure what you mean when you say I am referring to "mathematical machinations" and "wave analysis" because I don't think about sound propogation that way because I can't. Wave analysis is definately not one of my strong points. But a oscillation is an oscillation is an oscillation, however you want to call it.

 

[crashsite]

So how would you look at something that's moving back and forth? Or what would you call it?

Oh, there's oscillation involved and, I've agreed that, at some level, the need for analysing waves (balls and springs) is needed. But, you need to be careful of when you should be analyzing waves and when you should be analyzing something else. The oscillation and wave part is the gigahertz range oscillations of the molecules due to thermal activity. The vector effects of the collisions of those molecules oscillating at gigahertz rates as influenced by low rate disturbances is what's needed to understand sound propagation.

At least that's my take but, I welcome hearing how wave analysis at the low frequencies of sound generation cause sound to propagate.

Because that's what's happening. A lower frequency means the air molecule is moving back and forth over a greater distance. Nothing schmoozey about that.

I don't believe that for a minute and, I think that if you think about it, you really don't either.

I'm not sure what you mean about "wave analysis" though. Because I have a pretty particle-like mindset about sound.

Wave analysis being the belief that sound is propagated (somehow) by the interchange of potential and kinetic energy over times associated with audio (and similar) frequencies.

 

[dknguyen]

At least that's my take but, I welcome hearing how wave analysis at the low frequencies of sound generation cause sound to propagate.

...

Wave analysis being the belief that sound is propagated (somehow) by the interchange of potential and kinetic energy over times associated with audio (and similar) frequencies.

Remember that animation I sent you that's how I think about sound propogation. Not that transfer of energy stuff. How is it any different at higher frequencies than lower ones?

I don't believe that for a minute and, I think that if you think about it, you
really don't either.

Explain to me what I believe then and why you don't. I once again point back to that animation which clearly shows how
longer wavelength = molecules move back and forth a larger distance = more distance between the high pressure peaks where all the collisions (or repulsive forces as you say) are happening. I don't see what's not obvious about that.

 

[crashsite]

Waving Goodbye to Waves

Remember that animation I sent you that's how I think about sound propogation. Not that transfer of energy stuff. How is it any different at higher frequencies than lower ones?

I assume you're referring to the tuning fork oscillating and, boy by golly sending out those waves and then the picture of the pressures as relates to the sine wave and the obvious wave nature of it all?

How it's different at higher frequencies can be sort of analogized by thinking of RF modulation. Both the RF and the modulating signal are electrical. The RF serves the function of efficiently radiating the signal, as an electromagnetic signal, over a long distance. The audio (reference audio since we're talking about sound) is also an electrical signal but at frequencies that carry the payload information but, do not radiate very well. But, when it's impressed onto the carrier the whole package can be sent with near the efficiency of the RF itself.

In the case of sound, I believe that the thermal gigahertz rate oscillation is mechanical and the impressed audio signal is also mechanical. The sound propagation is then, also mechanical; moving from molecule to molecule by collisions. The speed of the propagation is based on the speed of the thermal movements of the molecules.

I know that seems a bit convoluted and takes a little effort to get your mind around (at least I'm finding it to be so). Things are complicated even further when you must consider that the audio disturbance, itself, consists of molecules that are also vibrating at gigahertz rates.

Explain to me what I believe then and why you don't. I once again point back to that animation which clearly shows how
longer wavelength = molecules move back and forth a larger distance = more distance between the high pressure peaks where all the collisions (or repulsive forces as you say) are happening. I don't see what's not obvious about that.

The whole concept of wavelengths in space is meaningless when the issue is ound propagation. Pressure gradients out there in the air are merely artifacts of the sound propagation process.

To think about sound propagation you need to clear your mind of the wave nature of sound altogether and that's what classically trained physics types just can't seem to do. You need to think about sound propagation in terms of near instantaneous time increments (the time it takes for one molecule to nudge another) and let the wave artifact chips fall where they may.

There's plenty to learn and plenty I don't understand about how those molecules act and interact to make the sound propagate. But, I feel like I am on the right track and am slowly making inroads and could use some help from people who can get that damn wave analysis out of their thinking (or, conversely, convince me that I'm thinking wrong and that analyzing waves yields the answer).

 

[crashsite]

Keep a steady hand on the tiller and steer true, me bucko

I was wrong there.

As the air pressure drops, the speed of sound will also be reduced meaning the waves will get longer.

It sure seems like that should be half true. Empirically, it seems like less air density would give a slower speed of sound but, a slower speed of sound would give shorter waves for a given frequency. But, that's moot because the problem is that basic premise is absolutely NOT true.

My gut feeling is still telling me that higher frequencies will be attenuated more but I can't explain why. My inner voice is failing me.

I'm replacing my original reply which was probably too pessimistic. Let me take a shot at it.

The traveling and standing waves in space may be artifacts of the sound propagation process but, they are still real and, they can be analyzed as waves because they are waves. Let's take the case of a 10 kHz signal. At a propagation speed of 1100 feet per second, one wavelength is about an inch.

Now, air is light and mobile as compared to liquids and solids. The molecules can be blown in the breeze and be affected by thermals, etc. So, how far does a 1 inch wave have to travel before air movements shift those molecules by half an inch? in turbulent air, maybe not very far. In still air, further. A shift of half an inch can put a 10 kHz wave 180 degrees out of phase with itself, essentially destroying it.

With normal air movements, those phased attenuations will average out to some overall attenuation (which is almost always also varying).

If the signal is 1 KHz, that same half inch of air movement will only affect the signal by a few degrees and thus there will be less nulling action and less attenuation.

But, here we're talking about the waves and waveforms and not the process of the propagation of them. There's a time to analyze waves and a time to analyze something else.

 

[dknguyen]

I assume you're referring to the tuning fork oscillating and, boy by golly sending out those waves and then the picture of the pressures as relates to the sine wave and the obvious wave nature of it all?

That sounds just like Jasonbe asking why electrons (or anything for that matter) accelerate back and forth in a sinusoidal matter instead of a square wave (instant-stop at the extremes with constant speed in the middle).

No, I am talking about this animation.**broken link removed**

The whole concept of wavelengths in space is meaningless when the issue is ound propagation. Pressure gradients out there in the air are merely artifacts of the sound propagation process.

To think about sound propagation you need to clear your mind of the wave nature of sound altogether and that's what classically trained physics types just can't seem to do. You need to think about sound propagation in terms of near instantaneous time increments (the time it takes for one molecule to nudge another) and let the wave artifact chips fall where they may.

I don't know why you keep referring to wave nature. I don't think about sound like that. I think about it as molecules nudging up against each other, not about kinetic and potential energy transfers. You also brush off pressure as an artifact which it's not. Artifacts are things you can get rid of when working in theory. Does sound propogation cause the pressure or does pressure cause the sound? Chicken or the egg? It's irrelevent because one cannot occur without the other, even in theory. A similar thing also happens when you try to figure out how a wing makes lift. To me it's not very different from diffusion or why a gas will flow out to fill a vacuum. Things close to each other that exert repulsive forces are bound to shove against each other harder and thus spread out again. I suppose you have a beef with how diffusion works too?

In the case of sound, I believe that the thermal gigahertz rate oscillation is mechanical and the impressed audio signal is also mechanical. The sound propagation is then, also mechanical; moving from molecule to molecule by collisions. The speed of the propagation is based on the speed of the thermal movements of the molecules.

Those random thermal kinetic motions due to temperature determine how far the molecules tend to be spaced out in the gas. And then you superimpose those vibrations on with the net movement caused by the vibration source. Like AC noise (thermal) superimposed on DC (the sound itself). Without those thermal vibrations, the air wouldn't be dissipated like a "gas". It would kind of be clumped up here and there- more like pool balls just sitting there. Nothing convoluted about that. I thought that was a given and no major breakthrough?

How are you visualizing the thermal motion of the molecules? Are you visualizing the motions as small relative to the particle? Or large? I'm seeing them as large on the same scale as the motion due to the sound disturbance itself. Not a molecule just sitting on the spot vibrating like a car in idle (or even a car on the highway). More like a drunk driver trying to stay in the lane. It's also a fluid, not a bunch of particles so every molecule is able to influence all the molecules around it.

 

[crashsite]

Getting there.

No, I am talking about this animation.**broken link removed**

Sorry, picked the "wrong" one. But, no matter. Same problem. The notion that sound propagates by wave action. Waves propagate by wave action. sound propagates by a completely different mechanism (as you seem to grasp further down...making me wonder why you are so adamant that I look at wave animations).

I suppose one could try to think of the animation, not as a macro of a wave but, rather as a micro of molecules washing outward in a wavelike manner. I think either view is problematic. You need an animation that puts a vector bias on randomly moving molecules.

I don't know why you keep referring to wave nature. I don't think about sound like that. I think about it as molecules nudging up against each other, not about kinetic and potential energy transfers.

I don't want to keep referring to wave nature but, it seems like everybody else doesn't want to talk about anything except waves. I want to get away from waves and think in terms of instantaneous actions.

You also brush off pressure as an artifact which it's not. Artifacts are things you can get rid of when working in theory.

I define an artifact as something that's either created or is a residue of some other process. Sound is generated by a distrubance and that disturbance is propelled through a medium at Mach 1. Anything else is an artifact. That there is a traveling or standing wave or pressure gradients is an artifact of the process.

That's not to say that the artifact is unimportant or peripheral. In fact, when it comes to sound, it's pretty much the purpose of it all. Doesn't change the fact that it's an artifact of the process of sound propagation.

Does sound propogation cause the pressure or does pressure cause the sound? Chicken or the egg? It's irrelevent because one cannot occur without the other, even in theory.

A disturbance occurs in a pool of randomly moving molecules and that initiates the process of sound propagation. How you view the order and speed that things happen or how you utilize or measure them determines how you think of them, I suppose.

I suppose you have a beef with how diffusion works too?

Those random thermal kinetic motions due to temperature determine how far the molecules tend to be spaced out in the gas. And then you superimpose those vibrations on with the net movement caused by the vibration source. Like AC noise (thermal) superimposed on DC (the sound itself).

No problem with any of that except to note that some of the stuff I still need help learning is the nature of the vibrations and how they are constrained within the medium. Factors that I believe are crucial to understaning sound propagation.

Without those thermal vibrations, the air wouldn't be dissipated like a "gas". It would kind of be clumped up here and there- more like pool balls just sitting there. Nothing convoluted about that. I thought that was a given and no major breakthrough?

You've sort of described the Bose-Einstein condensate. Glad to see you employ the pool ball analogy I had posted way back when.

How are you visualizing the thermal motion of the molecules? Are you visualizing the motions as small relative to the particle? Or large? I'm seeing them as large on the same scale as the motion due to the sound disturbance itself. Not a molecule just sitting on the spot vibrating like a car in idle (or even a car on the highway). More like a drunk driver trying to stay in the lane.

I don't know how to visualize it. I guess I envision the vibrations to be large enough relative to the spacings of the molecules as to allow them to collide. Or, at least heavily interact in a manner similar to colliding.

But, more important, I think, is how those vibrations are constrained such that, with a given temperature, they can have a given speed...independent of the pressure (ie: how far apart they are). Or, if they don't have a given speed, how they can impart a fixed propagation speed of Mach 1 to a disturbance...again, independent of the pressure.