Jump to content

Understanding and Optimizing Tactile Feedback


dominguez1

Recommended Posts

Continued from another thread...

 

 

Tactile feel is determined by sound field properties.

Without intensity a sound wave can not have a physical effect on anything, and for intensity you need pressure and velocity.

What is less clear is exactly what properties are important in which frequency range, and how the relationship between sound field properties affect how we experience sound.

 

Unless others beat me to it, I will post examples later, on both subjects.

 

 

Interested in your thoughts here on the SIL/PVL discussion. This is not what I've observed, but would certainly like to hear a different perspective.

 

This is how I understand tactile feel as well.

 

Here is a test I performed using an accelerometer to measure tactile feel close to the subs, and further away from the subs using a 15hz sine wave. This test demonstrates that SPL alone is not the sole contributor to tactile feel. Give this equation: Sound Intensity = pressure * particle velocity, the test demonstrates in this example that PVL must be the differentiating factor to the variation in tactile feel.

 

http://www.avsforum.com/forum/113-subwoofers-bass-transducers/1488059-your-home-theater-ulf-score.html#post23676269

 

 

@SME, if you look at measurements of different system, I think you will find that even for those with a reasonably flat and smooth frequency response, there will still be deviations in the phase response.

 

If both phase and freq resp is flat, that would indicate that the impulse response is also covered.

Practical systems consist of parts that are not necessarily put together in such a way that the total response is a minimum phase system.

 

For information on what we already know about tactile feedback, there are several thread on avsforum covering the subject.

I believe there were some good point made in the bossobass-thread as well.

 

Tactile feedback is not simply about feeling the bass in your chest, or shaking the coach.

At the utter extremes, imagine headphones compared to a good, full-frequency range, full-capacity loudspeaker system.

 

On headphones, you can hear the bass, but there is no point in turning up the volume, especially at the lowest frequencies all you achieve is this pressure at the ears.

 

On the speaker system, however, that is a different world.

The very lowest frequencies can be felt as movement - from the floor, sofa, and perhaps pressure sensations in the ears.

As you increase the frequency, you feel the bass like a slightly subtle push on your whole body.

(It is this push that turns the music into a feeling of sound-waves that kind of moves through your body when you turn it up - the louder, the more insane it gets.)

Then around 30-50hz you get a solid punch from the biggest drums, and this sensation from transient attacks continues further up in frequency, 80-120hz is often mentioned as the chest-impact-range.

 

Perhaps others could fill in on my attempt to describe this tactile feel, it is not always easy to describe sound sensations, and others may have different experiences.

 

 

I think nearly all 'tactile feel' is mainly resonant in nature, unless SPL is high enough to literally be hit by a 'blast wave' of bass.  Chest and abdominal cavities, furniture, floors, walls, all contribute to the sensation to a certain extent.

 

Dom,

 

How did you isolate the accelerometer to do your measurements?  Concrete pier or cinder blocks?  If the accelerometer was on a piece of furniture or anything else that could resonate at the freqs tested, you may have just been measuring sympathetic resonances. 

 

 

 

JSS

 

 

Not only do I believe that SIL/PVL are not relevant, but I believe there are far better explainations for why tactile preception differs between systems and why near-field placements often provide tactile perception for the same SPL.

 

Tactile preception of vibration and sound is a very complex subject, one that is probably even more complex than hearing.  Here are some lecture notes on the subject of tactile sensation including vibrotactile sensation.  Our body is capable of sensing vibrations in different parts of our skin, in our muscles, and even in some of our internal orgrans.  We can also hear vibrations in our bones that are transmitted directly to our ears.  Each of these sense modalities is integrated along with our hearing to form a perception.  Each sense likely contributes a different character to our perception, but at the same time sensation via one modality can mask sensations from other modalities.  All of these things affect what we hear and feel in very complex ways.

 

One thing we should all be able to agree on is that tactile perception necessarily requires vibration to be transmitted to the skin or through it deeper into the body.  Vibration is most efficiently transmitted into the body via direct contact with a vibrating solid surface (or liquid surface, if you like to get your bass on in the swimming pool), but transmission directly from the air to the body is also relevant.

 

Let's consider vibration transmitted through solid surfaces first.  The two surfaces that are most likely to transmit vibration to you are the floor, if your feet are in contact with it, and your seat.  All else the same, the stronger these vibrations are, the more bass

you will feel.  One transmission path for these vibrations is between the subwoofer, the floor, your seat, and your body.  Most subwoofers produce mechanical vibrations in addition to sound because they contain moving masses.  It is possible to reduce or eliminate this transmission path by using a dual-opposed woofer configuration or by mechanically decoupling the subwoofer from the floor.  It can be difficult, however, to mechanically decouple the lowest frequency vibrations between the sub and the floor.  Even if the subwoofer is perfectly isolated from the floor or avoids vibrating entirely by using a perfectly balanced dual-opposed configuration, vibration will still be transmitted to the floor by the sound itself.  Generally, the higher the SPL of the sound, the more vibration will be transmitted, so the greatest transmission will happen at the edges and corners of the room and especially at any surfaces near to a woofer.  A balanced or decoupled down firing woofer will still transmit plenty of energy to the floor simply because part of the floor is very close to the woofer where the SPL is very high.

 

Now let's talk about why you are more likely to feel vibrations with a sub placed close by than one farther away.  First of all, the level of vibration in the floor and other room surfaces decreases with distance from the source.  This is true of the sound coming from the sub also.  However, whereas the sound is constrained by your room boundaries vibrations in the floor and walls are free to travel to adjacent parts of the structure.  For this reason, it can be expected that the reduction in vibration level with distance for a floor is more substantial than the reduction in SPL with distance through the air.  Another consideration is damping.  Damping is the phenomenon by which mechanical energy is dissipated as heat.  Air damps sound very poorly except at the highest frequencies.  Damping in solids on the other varies by material, and some materials can damp bass energy fairly well.  The reason damping is important is because it reduces how much vibration is transmitted through solid surfaces over a distance.  If your subwoofer is located relatively far away, then the vibration that is transmitted into the floor around your subwoofer is dissipated a lot more before it reaches your feet and your seat than it would be if your subwoofer was located close by.  I believe these are the most likely reasons that close sub placements yield more tactile stimulation.

 

If you listen with your feet off the floor and you mechanically decouple your seat from the floor, you may be able to further reduce the amount of vibration reaching your body.  However, vibration will still transfer directly from the air to the seat itself.  Suppose you instead listen while standing up on a very heavy and firm surface (like the earth), then you will only feel vibrations that are transmitted directly between the air and you and your clothes.  This can and does happen, and its impact is not at all limited by the distance of the transducer to the listener, as long as the SPL is high enough.  Most likely this is the phenomenon at work for JSS when he felt the "mule kicks" at the movie theater in Austin.  I personally have felt some great bass kick from large scale outdoor sound systems at distances of 1/4 mile or more.  I think all of us have felt the bass from fireworks even from miles away.

 

I have a hypothesis that this air-to-body vibratory transmission is most impressive when the response is fairly flat, particularly in the 60-120 Hz region.  (A slight bump there relative to the other frequencies seem to hurt much.)  As I already aruged on the BossoBass thread, my hunch is that the flatter response delivers the energy of a transient more compactly in time leading to a higher peak SPL and a stronger perception.  I may be wrong here, and I will admit that I have felt quite a bit of "punch" from e.g., ported systems with excessive ringing at their port frequencies that otherwise sound like crap.  On those ported systems, just about anything that punches feels about the same and makes a similar drawn out "oooooph" sound as it happens.  On the other hand, while flatter systems may not hit as hard, they seem faster, they differentiate different sounds (i.e., different kick drums), and they are more likely to kick in general.  With a nice flat system, the kick is often felt strongly even though the sound is barely heard.

 

So if my hypothesis about a flat response is correct, then another good reason that close sub placement yields better kick is because it's response in that critical 60-120 Hz region is likely to be more smooth.  The response is likely to be more smooth with the close placement because the direct sound is (relatively speaking) greater than the reflected sound from the boundaries.  Unfortunately, close sub placements often perform worse at lower frequencies because the wavelengths are so large that the room response overwhelms.  Unless you can put your ears within inches of the woofer, you are still hearing mostly the room at the lowest frequencies.  For that reason, I use mid-bass subs placed close to my seats in conjunction with dedicated deep bass subs located at the front of the room.  I actually get *stronger* tactile response in the deep bass with my subs placed at the front of the room than I did when they were placed close by because the sub/room coupling is more efficient overall.  For me, the setup use gives me the best of both worlds.

 

Now, let me go back to the subject of vibrotactile sensation via transmission from the seat or floor and the issue of non-flatness of this response.  At one point in the past, I experimented with installing decoupling feet on my close-placed mid-bass subs.  I chose a design that isolated most vibrations above 50-60 Hz or so, which is close to the end of their range.  After installing the feet, I was able to roughly verify that they worked as designed by doing before and after SPL measurements.  After installing the feet, I lost about 1 dB SPL overall starting around 50-60 Hz, presumably because the floor was no longer radiating this extra energy into the air.  The subjective change that resulted from doing this was quite dramatic.  I definitely noticed a lot more of the "chest kick" type of feeling I associate with large scale sound systems where I am standing on concrete and too far away from the transducers for solid surface vibrations to significantly contribute to the sound impression.  My experience strongly suggests that the vibrations felt through solid surfaces can overwhelm and mask and those felt through the air, and that the poor transient response of the solid surfaces can substantially degrade the quality of the overall perception.  However, I also did not prefer the sound in the end result, nor did my wife.  I thought the bass sounded very quiet, distant, and poorly defined.  Turning up the sub level 6 dB brought the subjective level back up, but it still sounded distant and poorly defined.  What happened here?  It took me a while to work it out, but I believe the biggest reason we preferred the decoupled sound less is because my in-room upper bass response was very poor with severe dips throughout.  What I was hearing with the decoupling was a total loss of upper bass, even though the measured SPL response only dropped by 1 dB.  The vibrotactile sensations from the floor and sofa were contributing to the preception of the audio and filling in the dips of my in-room response.  It was quite remarkable to realize that much of what I thought I was hearing, I was actually feeling, but my brain was melding the perceptions together.  Now that I've added bass traps and have ditched Audyssey for my own EQ solution, my upper bass response is worlds better.  I still have a ways to go, but once I'm satisfied with how flat it is, I'll probably give decoupling the mid-bass subs another shot because it really did help with the chest kick sensation.

 

From this exercise, I learned that the importance of vibrotactile sensation to bass perception cannot be ignored.  In the vast majority of listening rooms these sensations likely make a considerable contribution both to what we hear and what we feel.  As such, I believe that controlling and equalizing vibrotactile response is an important frontier in the endless quest for better audio.

 

 

I'm sorry, but I don't see any evidence in that test that sound intensity has anything to do with tactile sensation.  You have most certainly demonstrated (even using the rather crude accelerometers in your phone) that solid surface vibrations are increased with your sub placed close by.  I give a very reasonable alternative explaination above for why this occurs.  There is no need to consider the sound intensity or particle velocity to understand this.

 

 

JSS,

 

The accelerometer was placed directly on the couch.

 

If I placed the phone on the concrete floor, there would be little doubt that it would not produce any significant reading whatsoever over the noise floor of the app.

 

You are 100% correct in that the couch's resonant frequency was the cause for the vibration. However, that vibration increased closer to the source, given the same SPL.

 

SME,

 

Thanks for your detailed response.

 

You mention that "the level of vibration in the floor and other room surfaces decreases with distance from the source". The question is: what's properties of sound is causing the vibration? It is not SPL alone as my test demonstrates; the SPL of both sub sets were calibrated to ~96db. 

 

My room is on a concrete slab. The couch sits on the concrete. At 96db @15hz, the concrete is not vibrating in any appreciable amount at all and is not transferring any vibration to the couch.

 

If the couch is not being influenced by the concrete vibrating, and the SPL is the same for both distances, then why is the couch vibrating more in the scenario where the sub is closer to the couch?

 

 

Hmm.  Good question!  How close to the couch is the sub?  If the sub is placed adjacent to the couch, as mine are to mine, then part of the couch is probably seeing a much higher SPL than your listeners are.  So the vibrations are being transmitted from your sub to the air to the couch adjcent to the sub and then to your torso and butt.  I still don't think SIL/PVL has anything to do with it, apart from the extent that they correlate with SPL.

 

 

Link to comment
Share on other sites

  • Replies 88
  • Created
  • Last Reply
 

 

The subs are about 1 foot away from the couch and ~3.5ft away from the main LP.

 

I didn't measure a foot away from the couch, but if I did, the SPL would more closely resemble the close mic SPL of  the sub. The close mic output would typically be less output than the output further away because the measurement further away would account for room gain, where as the close mic response would be more room gain agnostic.

 

One thing you haven't quantified in your hypothesis is what you're calling "vibrations". These vibrations must be excited by the sound wave. My tests prove that SPL alone is not the sole contributor, so it must be something more...which leads us to the Sound Intensity hypothesis.

 

My test was a simple one.

 

If you rule out the vibration transfer function because of the concrete slab, and calibrate SPL as the same, it must be some other sound property. Given that physics shows SIL = SPL * PVL, it must mean the PVL would be the differentiating factor in this example.

 

I would like to hear alternative views that refute this hypothesis, but have yet to been provided any supporting evidence of the contrary.

 

Is there any additional insight that you can provide as to what's causing this? There aren't any other levers to pull that I'm aware of, which leads us to supporting the SIL hypothesis.

 

 

One factor that I believe is being missed is that your SPL measurements for matching the 2 systems output are taken at a single point in space at the microphone element and in no way take into account the tactile vibrations which are the actual function being looked at. The phone was being set on the couch for the vibration measurement...The couch is a very large piece of furniture occupying a significant volume of space in the room relative to a microphone. In effect the entire couch is the microphone at that point. When you sit on a couch you are now in contact with a very large surface area of the couch. The subs were in very different parts of the room and affected by the room reflections and boundaries much differently. The total amount of energy from each sub system physically imparted into the whole of the couch as vibration will be much much different. This explains the difference in tactile sensation. The near-field sub would have a much more focused and directed energy on the couch than a sub 15 to 20ft away. The 2 subs may have had the same SPL at the spot measured but what about back behind the couch in front of the near sub? Near the floor on the left front corner of the couch, what about the back right corner? Under the couch in the middle? That's what I'm saying you would need hundreds of SPL measurements and a few accelerometer placements to get anywhere close to understanding the total energy impacting the couch from each sub system and calibrating them to each other.

 

A single point SPL measurement where the phone will be placed on the couch simply does not have the resolving power to look into the vibration issue.  In a complex acoustic environment with direct physical contact with other objects you will have the audible component of sound and also the physical sensations. The two are separate issues to a large extent on that we agree. However I'm not yet convinced that it has anything to do with particle velocity or sound intensity.

 

 

Let's define the energy that you're referring to.

 

Sound Energy is a form of energy associated with vibration or the disturbance of matter.

 

Sound Power is sound energy per unit time.

 

Sound Intensity is sound power per unit area

 

Therefore, Sound Intensity can be said to be sound energy per unit time/area.

 

Is Sound Intensity not what is causing the vibrations in the couch? I'm trying to understand why SIL would not be a factor?

 

 

Sound intensity and sound energy are often very closely linked with SPL (dB) and each other.

 

The entire point of my post was that the "matching" of spl of the 2 subs at a single point in a vast 3 dimensional, complex acoustic space, while having both subs in radically different positions within that space, is a data point of very little relevance for the exercise being undertaken. It simply doesn't have much at all to do with the amount of total energy and vibration being put into the couch by the individual speakers. You can call that energy whatever you want, but the point is that to imply that the SPL levels surrounding and impacting the couch were level matched between the subs so that regular old sound pressure level is removed from consideration as a cause of the difference is not correct.

 

Move the mic 3 feet to the left and the SPL will not be the same between the 2 subs.

 

 

If you place a subwoofer very close to a large coach, the coach will also change the soundfield because it is large and too close to the source.

 

However, @dom's experiment shows an attempt to verify an observation by objective measurement - the coach vibrates more with the subwoofer close up.

It is easy to show mathematically that the soundfield has a larger intensity and a larger velocity, relative to pressure, close to the source. 

 

Practical experiments as in attempts to make an overall improvement in sound quality seems variable.

Some think it is awesome - the coach vibrates more, some find there is no overall improvement.

 

I think this varies a lot depending on room, system configuration.

 

I recently did an experiment with this - placing two subwoofer units very close behind the sofa.

The best result was with units running out of phase with tuned delay.

It did not add any good sensational contribution, as the sofa vibrates more, but not in a good way, simply more of the wrong vibrations.

This is of course very dependent of the construction of the sofa, which will determine its resonance pattern.

It was impossible to tune the complete system into a decent impulse response, so the overall sound was very compromised compared to the current normal set-up.

 

 

I agree. What I am saying is for the purposes of that experiment there is an elephant in the room that has been overlooked from what I have read so far.

 

Let us say that you have one subwoofer directly behind a couch about 1ft away and firing directly into the back of it. You also have another sub 15ft away from the couch in a front corner. Now lets say that you place a microphone to measure the SPL from each at the headrest position of the middle seat on the couch. You run a 20Hz tone through each sub and set the recorded SPL of each to 100dB. What happens when you move the microphone 4ft to the right and sit it on top of the right corner of the couch and recheck the levels of the two subs? Will it be exactly the same? Most likely not. We can't know for certain without actually measuring it. What if we move the mic all of the way across the back of the couch to the opposing position, or place it near the floor in the front of the couch? We cannot say how the 2 subs relative levels will change without measurements.

 

Now in this same scenario let's say we then move the mic back behind the couch directly in front of the sub firing into the back of the couch and again measure the relative SPL levels. I think we can reasonably guess that the close sub is going to have much higher SPL levels relative to the corner loaded sub across a large portion of the back of the couch. Causing the couch to vibrate more. Yes it will also exhibit higher intensity, PV differences and all that as well, but the SPL is also much higher and more evenly distributed from the near field sub into the back and probably bottom of the couch. We cannot reasonably deduce which effect is causing this at the exclusion of everything else (regular old SPL) with this type of experiment IMHO. There are too many variables.

 
Link to comment
Share on other sites

Josh, I understand what you are saying, but just wanted to get on the same page WRT Sound Intensity.

 

By definition, SIL is in play here. It started to sound like it was a mythical property, so just wanted to define it.  ;)

 

Don't have a whole lot of time at the moment, but will post a counter view soon.

Link to comment
Share on other sites

Looking at Josh's last quoted post above, I find it impossible to argue against the point being made.

 

100dB @ 15 Hz at point A may be 110dB @ 15 Hz at point B for near field sub and 95dB at point B for far field sub, and the vibration of the whole couch is being measured. That's a compelling argument.

Link to comment
Share on other sites

Hi Dom,

 

Have you thought of investigating only PV by putting both of your subs 180 degrees out of phase with each other and comparing the physical sensation/vibration with them far and then nearfield?  Wouldn't that take SPL almost out of the equation?

 

The phone on the couch thing...

 

The couch is acting as a buffer and it will likely react with some frequencies while damping others more than say a table or kitchen chair will.  Maybe I'm wrong, just my thoughts about it. 

Link to comment
Share on other sites

Josh, Dave, et al,

 

I am traveling this week, but will take additional SPL readings around the couch for both the nearfield and farfield placement.

 

My hypothesis is that the SPL variation will vary across the different points around the couch, but the average nearfield SPL will be very close to the average farfield SPL.

 

One point I will cite in my test; I had to increase the volume of the farfield subs by 20db to get a similar reading on the accelerometer. And even at that volume, it still did not exactly match the reading of the nearfield subs.

 

I point this out showing that there is a vast difference in tactile feedback wrt SPL between nearfield and farfield. So vast, IMO, that deems the precision of the additional tests around the couch unnecessary. I highly doubt that when performing these additional tests will show a 20db average lower SPL level for the farfield subs compared to the nearfield. If SPL is the only factor, then that's what the approximate difference I would need to see to show the difference in tactile feedback given the above, correct? Am I thinking about this in the right way?

Link to comment
Share on other sites

Hi Dom,

 

Have you thought of investigating only PV by putting both of your subs 180 degrees out of phase with each other and comparing the physical sensation/vibration with them far and then nearfield?  Wouldn't that take SPL almost out of the equation?

 

The phone on the couch thing...

 

The couch is acting as a buffer and it will likely react with some frequencies while damping others more than say a table or kitchen chair will.  Maybe I'm wrong, just my thoughts about it. 

Not sure I follow...if the both sets of subs were both active during the test, how could I measure which is contributing to the shaking?

 

To your second point, I thought this as well...essentially, the couch is 'absorbing' some SPL in the nearfield, and by removing the couch, SPL should increase. However, I did test this as documented in the example...when I removed the couch and remeasured the SPL at the same LP, the SPL remained at 96db.

 

So with or without the couch, SPL remained the same.

Link to comment
Share on other sites

I mean to bring them both in to nearfield and then both out to far with them out of phase with each other.  It will show how much PV varies (with distance) removing SPL as much as possible by putting the 2 all the way out of phase with each other.  Then do the same test measuring SPL.  Hopefully the test will show the relationship between SPL and PV with your rig at both far and nearfield.

Link to comment
Share on other sites

If you place a subwoofer very close to a large coach, the coach will also change the soundfield because it is large and too close to the source.

 

However, @dom's experiment shows an attempt to verify an observation by objective measurement - the coach vibrates more with the subwoofer close up.

It is easy to show mathematically that the soundfield has a larger intensity and a larger velocity, relative to pressure, close to the source.

 

Actually, it's not that easy.  The easy math arises from assumption that aren't necessarily valid.  For example, both the near-field and far-field approximations don't apply here.  Both consider a source (or sources) in the absence of reflections.  The longer the wavelength of bass, the more the room dominates the response.  Even 1 foot away may not be near-field at a low enough frequency, and by then you start non-spherical effects due to the fact that the woofer isn't a point source.  His measurements were at 15 Hz, which is very low as far as his room is concerned.

 

I recently did an experiment with this - placing two subwoofer units very close behind the sofa.

The best result was with units running out of phase with tuned delay.

It did not add any good sensational contribution, as the sofa vibrates more, but not in a good way, simply more of the wrong vibrations.

 

This certainly raises the ratio of vibration to sound, but not without great harm to the sound.  It's also a great way to piss off your neighbors.  Conservation of energy basically says that if the bass isn't going to your listening area then it's going somewhere you don't want it to go.

Link to comment
Share on other sites

Not sure I follow...if the both sets of subs were both active during the test, how could I measure which is contributing to the shaking?

 

To your second point, I thought this as well...essentially, the couch is 'absorbing' some SPL in the nearfield, and by removing the couch, SPL should increase. However, I did test this as documented in the example...when I removed the couch and remeasured the SPL at the same LP, the SPL remained at 96db.

 

So with or without the couch, SPL remained the same.

As far as 15 Hz is concerned, the sofa doesn't do much.  It probably absorbs some energy, but not a whole lot.  The sound at that frequency basically just goes around it.  In fact, what the sound really sees is more like a leaky pressure vessel.  This is not adequately modelled as being in either the near-field or far-field.  It needs a model of its own.

 

Let's define the energy that you're referring to.

 

Sound Energy is a form of energy associated with vibration or the disturbance of matter.

 

Sound Power is sound energy per unit time.

 

Sound Intensity is sound power per unit area

 

Therefore, Sound Intensity can be said to be sound energy per unit time/area.

 

Is Sound Intensity not what is causing the vibrations in the couch? I'm trying to understand why SIL would not be a factor?

Out of curiosity, I reviewed the concept of sound intensity among other things.  There is some confusion here because in the texts, sound intensity is usually considered as a quantity within a homogeneous (i.e., same throughout) medium like the air.  Thinking about it, the concept of sound intensity can also be applied to quantify the amount of energy being transmitted at the interface of two dissimilar media, such as the air and your couch.  What you will find, however, is that this value is very different from the sound intensity you would see in the air away from boundaries.  It is constrained by the fact that both the pressure and the particle velocity in the direction normal to the interface (perpendicular in both dimensions of the surface) have to be continuous.  By continuous I mean, they can't suddenly jump in value between the two media.  Instead of computing the sound intensity at this interface, what is more commonly done is to compute a reflection coefficient (a complex valued coefficient representing both a magnitude and phase) which describes the ratio of reflected sound intensity to incident sound intensity.  The difference between these two is the amount of energy transferred.  To calculate the reflection coefficient one typically determines the acoustic impedances of each medium across the interface.  The acoustic impedance is the ratio between pressure and particle flow rate through a surface.  If the particle velocity is uniform, then the particle flow rate is just the particle velocity times the area.  The closer the impedances are, the more energy that is transmitted instead of reflected because the particles tend to respond more similarly to changes in pressure.

 

I wouldn't assume your concrete slab is insert.  Generally speaking, the impedance of solid materials is much higher than that of the air.  Something like the concrete slab has a very high impedance.  The legs of your sofa likely have a somewhat lower but still high impedance.  The cushioning material and your body likely have a moderate impedance.   An interesting point about this is that it may be more efficient for sound to travel from the concrete slab into your body by way of the sofa than for it to travel directly to your body by way of your feet.  Also, unless your sub uses a dual-opposed configuration to suppress vibration in the subwoofer cabinet, it will produce a lot of vibration in addition to the sound, and this vibration may be transmitted with enough efficiency into your concrete floor to matter.

 

In all likelihood, the impedance for something like a sofa isn't going to change much with respect to location in the room.  The impedance of the air may vary somewhat by the frequency and the shape of the room it has to fill.  For higher frequencies, the far field approximation is usually appropriate, and one can treat the impedance of air as constant, equal to its characteristic impedance.  As I stated already, I don't believe 15 Hz in your room can be adequately analyzed using either a near-field or far-field approximation.  In my recent reading, the closest model I've come across is the impedance tube.  The impedance tube is unfortunately 1D, so it doesn't account for the 3D spatial spreading of energy from the sub.  Nevertheless, for a wave whose length is significantly more than half of the length of the tube, the air closer to the transducer is actually lower in impedance than the air at the far end.  Because you want the air impedance to be higher so as to better couple with the solids that have much higher impedance, the implication is that you should actually try to listen away from the transducers.  More importantly, you should listen as close to a room boundary or corner as possible because that's likely where the air impedance is highest.

 

That's just one way of approaching the question though.  At 15 Hz, a lot of energy is getting transferred to surfaces throughout your room, and much of that energy is may even be re-entering the room at a different location.  The best you can do is to try to identify the path that the vibrations are travelling.  This is easier said than done.  Realize that your accelerometer cannot tell you how much energy is being transmitted in a substance.  To figure that out, you'd need to know that material's impedance as well.  You may feel or measure very little vibration in the concrete, but you are basically measuring the particle velocity of the concrete.  Because of its high impedance, it can transmit a lot of energy without having a high particle velocity.

 

For what it's worth, I used to run my deep bass subs much closer to the listeners than I do now.  Since moving the subs to the front wall, I haven't noticed any sort of drop in feeling from frequencies in the 15-20 Hz range.  I do have a suspended wood floor which may or may not matter.  My SPL is a few dB higher with the subs up front, just because I have more headroom for more SPL and do my own calibration now.  Also, with the subs placed close to me, I had a deep hole in my response from 30-45 Hz or so.  I didn't know it at the time because I didn't have a measurement mic nor take the time to measure the long way with an SPL meter.  By having the sub away from both walls, I actually had a nasty pair of first-order reflections that created the response dip and made it non-minimum phase.  I did feel bass at those frequencies, at least as well as I do now with the subs up front.  I naively thought it sounded good at the time.  In reality, there was no comparison.  With the subs up front, not only could I hear the bass down there more clearly, but it sounded much tighter down there too.  And for all the sensation I lost in my butt, I gained a lot of "lower gut punch".  What I lost was the upper chest punch along with mid/upper bass clarity.  I solved that issue by adding mid-bass subs close by to achieve the best of both.

Link to comment
Share on other sites

I mean to bring them both in to nearfield and then both out to far with them out of phase with each other.  It will show how much PV varies (with distance) removing SPL as much as possible by putting the 2 all the way out of phase with each other.  Then do the same test measuring SPL.  Hopefully the test will show the relationship between SPL and PV with your rig at both far and nearfield.

I did a similar test except I ran my nearfield and farfield subs out of phase here. This test was at 27hz. There was a difference measured on the accelerometer. But that's not why I'm bringing this up.

 

Initially, I ran the same test at 15hz. No matter how far in our out of phase they were, I could only generate a 1db difference in SPL; whereas at 27hz I could generate a 16db swing. Why was that?

 

Coolrda sourced posts from Seaton and LTD here that I believe explains this phenomena:

 

I'm wrong on the 1/4 wave. The correct answer is 1/2 or less then half a wavelength. Heres a post explaining it by LTD02 on one of my builds from three years back.

 

 

 

"My rooms dimensions are 18ft x 19ft x 8ft."

 

"I measured room gain by comparing the Winisd graph against actual in room via the OmniMic and RG starts at +1db@20hz and increases at 10db per octave. It's +11db@10hz and +21db@5hz. Room gain is more than I expected, but I wish it started an octave higher."

 

pressure vessel gain kicks in when the long dimension of the room is less than 1/2 the wavelength at the critical frequency.

 

long dimension of the room = (H^2 + W^2 + D^2)^0.5 = (8^2 + 19^2 + 18^2)^0.5 = (64+361+324)^0.5 = 749^0.5 = 27.37 ft.

 

27.37 is the half wavelength, for the full wavelength multiply by 2 = 54.74 ft.

 

1130(ft./s) / 54.74 ft. = 20.64 (1/s) = 20.64 hz

 

the theoretical gain is 12db per octave, but that is if you are losing nothing to wall flexing, etc.

 

so your room is performing pretty much exactly according to theory.

 

 

Using a frequency/wavelength calculator my PVG with a long dimension of 15'4"X11'x9' starts around 22hz, I think which shows on the graph. This is where room gain begins and where phasing's effect ends. I have about the same PVG as room 1(19x18x8) in my current room, room 3{15'4"x11'x9'). I believe room 2 measured the same as well. So while TA/phase/distance can have a drastic effect above PVG it has none below. An average rooms PVG start point is around where ULF starts as well. I've seen this for years looking at the graphs but just never thought of the correlation of ULF and Phase. I think this is why Bossobass, years ago, always recommended sealed sub systems in sealed rooms. The subs 12db an octave rolloff matches perfectly with PVG's 12db gains per octave theoretically giving one a flat in room response to 5hz or lower. You still have to eq someway those freqs above PVG's start point. Your 16db loss from time misalignment above your PVG start point has zero db loss below, hence no change to your vibrometer reading. Then when you compensate by raising the sub level +16db, ULF or below PVG start point will be thunderous. Sound waves are phase dependent only above PVG start point. That how I interpret the graphs. I may be way off base here.

 

Time alignment should have zero effect from the 1/2 wavelength of the long dimension and rise to to have max effect at frequency of full wavelength of long dimension and all freqs above, in effect, a zero null to max null at an octave above it. Those dimensions make up the room nodes that wreak havoc with our rooms. See above posts last sentence for disclaimer if i"m :confused:

 

Heres another post by Mark Seaton  about a decade back on another forum. 

 

 "There are changes below where a 1/2 wavelength won't "fit" in the room, usually defined by the longest dimension, top-to-bottom of opposite corners. Below this point you no longer see modal behavior, problems, or variations. What you do see is a gain in LF output for a given subwoofer output. So in fact the result is the opposite: More output, less distortion. In simplest terms you transition to a mode of pressure modulation, where you are basically modulating the pressure in the entire room. Remember that our ears react to air pressure modulations and it doesn't matter what causes the level of modulation, since the observer (us in this case) is exposed to the same changes in pressure as if they were outdoors."

 

The way I understand it, essentially the 1/2 wavelength needs to 'fit' in the room in order to have any significant phase interaction that affects SPL. At 27hz, a 1/2 wavelength fits in my room, whereas at 15hz, it does not.

 

If the above is true (also demonstrated by my phase test at 15hz between nearfield and farfield subs), then the test Shred described would not show much variation in SPL to determine the impact of PVL...me thinks anyway. :)

 

Would like to hear thoughts on if I'm thinking about this correctly. If this is also true, in most rooms, the theory of mixing sealed and ported will have 'phase' issues at port tune, really isn't much of an issue.

Link to comment
Share on other sites

@SME, thanks for your additional insight...going to have to digest that one a while...but seems to be a reasonable alternative hypothesis. In some of your statements you mentioned particle velocity on the slab, etc....in getting more familiar with Sound Intensity, has your stance changed on the possibility of that being a differentiator?

Link to comment
Share on other sites

Actually, it's not that easy.  The easy math arises from assumption that aren't necessarily valid.  For example, both the near-field and far-field approximations don't apply here.  Both consider a source (or sources) in the absence of reflections.  The longer the wavelength of bass, the more the room dominates the response.  Even 1 foot away may not be near-field at a low enough frequency, and by then you start non-spherical effects due to the fact that the woofer isn't a point source.  His measurements were at 15 Hz, which is very low as far as his room is concerned.

 

 

This certainly raises the ratio of vibration to sound, but not without great harm to the sound.  It's also a great way to piss off your neighbors.  Conservation of energy basically says that if the bass isn't going to your listening area then it's going somewhere you don't want it to go.

 

I believe you made some points back in the other thread, so I am just going to comment quickly so that it doesn't look like I am ignoring input from you that is relevant and seems well thought.

 

I find it hard to disagree on the points you make here, this more or less matches my own view.

 

In a sealed room, at very low freq, the soundfield will approximate the same all across the entire room, and the velocity part will be low.

There will, however, be differences as you move in very close to the source, and more for increasing frequency.

Depends on the room, but I also agree that even very close to the source, in fact even inside a horn mouth, there is significant contribution from room reflections.

 

Here I have a measurement comparing nearfield with listening position.

The frequency response is very different, as expected, still we see that the nf is greatly affected by room.

The yellow line in the lp plot is the interesting one, it is the velocity measured at the lp normalised using the relation between spl and velocity for the nf.

At the lp, there is a huge loss in velocity.

post-181-0-66957400-1428521090_thumb.pngpost-181-0-97375700-1428521099_thumb.png

Link to comment
Share on other sites

@okv, I do remember you constructing a PVL probe from a 'shorted out' driver to be able to detect PVL levels. I assume this is the tool that you used (which is pretty amazing at least to give a general shape of the PV response) to create the below chart:

 

post-181-0-97375700-1428521099.png

 

How big is your room? What was the sub design in this test?

 

If I'm understanding correctly, the absolute db levels may not be accurate, but the relative db levels (or PV response curve; black) should be accurate, correct?

 

This appears to be behaving how I would expect a PVL response to behave as it transfers to different sound fields; meaning at the lower frequencies PVL is at its highest because the waves are in the nearfield sound field, and are out of phase with SPL. As the frequencies' length start to shift into the farfield sound field, PVL and SPL are more closely aligned and are in phase.

 

In the nearfield, one can not only consider SPL to determine Sound Intensity as represented by the shape of the PVL response vs SPL response being more different (especially from 18-35hz). However in the farfield, since SPL and PVL are in phase, one only needs to reference SPL to understand Sound Intensity. This can be seen by the response shape of both SPL and PVL around 35hz to 90hz as being more similar.

 

Thoughts?

Link to comment
Share on other sites

More or less correct.

 

This is just an example I found from a measurement, for the purpose of showing that the properties of the sound field changes a lot, also inside a room, as you move closer to or away from the source.

 

What we can see from the graph is that the velocity is lower in the listening position, and the difference is larger at lower frequencies.

This corresponds with theory.

We also observe that the velocity does not match the measured spl, especially at the null which is caused by the rear wall reflection, here the velocity is relatively large compared to spl. 

 

It is relatively easy to measure the velocity alone, you just need a velocity microphone, the principle is described in all electroacoustics textbooks, in the chapter on acoustic receivers.

If you intend to buy one for this purpose, you must make sure it works properly for low frequencies.

Measuring intensity is more difficult, and available solutions are not made for this purpose.

 

 

 

 

For some more information on the theoretical aspects of sound fields, here is an excerpt from something I am writing on:

 

How it works in more practical terms

In the near field close to a source pressure and velocity are 90 degrees out of phase.
Near field means that the distance to the source is small or comparable to the wavelength, the wavefront is spherical, so that the velocity potential kind of 'leaks' sideways.
In the far field pressure and velocity are in phase, as the wavefront now approaches a plane wave.

The fact that the particle velocity and the pressure both decrease linearly as a function of distance from the source means that the intensity will decrease with the power of 2 to the distance.
Close to the source, however, this situation changes.
Since the acoustic impedance increases for decreasing frequency close to the source, the velocity relative to pressure also increases, and the reactive intensity is larger.

Now, what does close to the source mean.
If near field is defined as d << L/2pi, we see that the term close depends on frequency, and that in normal small rooms we will always be in the near field at the lowest frequencies.

For a standing wave in a room the net intensity is zero, and pressure and velocity are 90 degrees out of phase.
The active intensity is zero, but the reactive intensity is high.

In a diffuse sound field - a room with lots of reflections - the net intensity is zero.
Reactive and active sound intensity are both zero.

In an active sound field where velocity and pressure are in phase, the amplitude of velocity relative to pressure is fixed, as the acoustic impedance is constant.
To achieve higher particle velocity amplitude it is necessary to create some kind of reactive sound field.

In all practical situations in real rooms reflections will affect both phase and amplitude of the velocity potential relative to pressure, so that intensity will be highly influenced by the room and quite different from the free field situation.
 

Link to comment
Share on other sites

I believe you made some points back in the other thread, so I am just going to comment quickly so that it doesn't look like I am ignoring input from you that is relevant and seems well thought.

attachicon.gifsi_1.pngattachicon.gifsi_2.png

Are your blue lines measured very near to the woofer?  If so, then it does look like that measurement locations gives a result pretty close to the near field approximation.  At the same time, as you move away, the room contribution increases disproportionately for the lower frequencies.  All of this makes intuitive sense.

 

What doesn't make sense is that if PVL drops relative to SPL when moving away from the woofer, then that would indicate an increase in impedance in the air, but that would imply more efficient not less efficient transfer of energy between air and solids because solids generally have much higher impedance.  This would suggest that the near field is actually bad bad place to listen if you want to feel vibrations.  The trouble as I said earlier, is that I think these are inappropriate simplifications.  Perhaps the near-field approximation does hold if you get close enough to the woofer, but once you leave that region, you'll never see anything resembling far field behaviour for wavelengths on the order of and larger than the room dimensions.  Once you get to the point that the waves are significantly bigger than the room dimensions, then the system starts to behave as a leaky pressure vessel.  In this case, I believe the leakiness of each surface will play a big role in what the sound-field in the room looks like.  This is a very difficult problem to solve because you need to know the mechanical properties of all the room surfaces in addition to the shape of the room interior.  What happens if we could seal the room perfectly?  As frequency drops, we'd see PVL essentially vanish implying a big rise in the impedance of the air itself and an increased tendency for the energy in the air to transfer to solids.  In other words, our attempts to perfectly seal the room are thwarted as the energy rushes into whatever object in the room has lowest impedance, which may be a sofa or body if the room boundaries are so solid.

 

That's very interesting, because it suggests that tactile sensation in the low frequency limit may be stronger in well-sealed small rooms than in leaky small rooms, which in turn may be better than much larger rooms and outdoor spaces.  The experiences of myself and others suggests that the chest resonance effect is more commonly experienced in larger rooms or outdoor environments.  To the extent that the chest primarily resonates with mid and upper bass where wavelengths are similar to and smaller than the dimensions of our typical indoor spaces, the sound field in the room will tend to be a lot more irregular with considerable impedance variations in space and frequency.  This field may be further modified by the properties of the boundaries.  Dry wall on studs and suspended flooring tends to pass energy in the mid-to-high bass frequencies quite well, so the air impedance may not vary as much in the room.  Whereas, with concrete walls and floor, very little mid-to-high bass energy escapes the room leading to a very modal response with considerably higher impedance in the "peak" regions of space and considerably lower impedance in the "null" regions.

 

The above is a bit speculative as I haven't worked through the ugly details on these kinds of problems before.  But no matter which way I choose to think about it, I'm inclined to believe that you want high SPL and low PVL for the most efficient energy transfer to solids.  These two conditions lead to high impedance, which improves coupling with solid surfaces and thus a high SIL *when measured at the interface*.

Link to comment
Share on other sites

But no matter which way I choose to think about it, I'm inclined to believe that you want high SPL and low PVL for the most efficient energy transfer to solids.  These two conditions lead to high impedance, which improves coupling with solid surfaces and thus a high SIL *when measured at the interface*.

Or the opposite: high PVL and low SPL. This is what is typically observed in the near field based on the articles I sourced in my original test (in the nearfield, PVL is greater). When in the near field, PVL and SPL are out of phase causing a wider gap (and resulting higher impedance) between the two compared to when they are in phase in the far field.

 

This also can be seen in okv's responses above.

Link to comment
Share on other sites

Are your blue lines measured very near to the woofer?  If so, then it does look like that measurement locations gives a result pretty close to the near field approximation.  At the same time, as you move away, the room contribution increases disproportionately for the lower frequencies.  All of this makes intuitive sense.

...

 

Around 0.5m, if you measure very close, as in 5mm, you will measure the velocity of the diaphragm, and since this is a compact horn unit, the total from the port and the woofer would not sum correctly.

 

The second blue line - velocity - is not the measured velocity, it is normalised using a transfer function created by dividing the spl on the velocity.

When using this same transfer function on the lp measurement for velocity, we get a velocity measurement that can be compared to the spl, so that we can see the difference between the near and lp situation.

Link to comment
Share on other sites

That's very interesting, because it suggests that tactile sensation in the low frequency limit may be stronger in well-sealed small rooms than in leaky small rooms, which in turn may be better than much larger rooms and outdoor spaces.  The experiences of myself and others suggests that the chest resonance effect is more commonly experienced in larger rooms or outdoor environments.  To the extent that the chest primarily resonates with mid and upper bass where wavelengths are similar to and smaller than the dimensions of our typical indoor spaces, the sound field in the room will tend to be a lot more irregular with considerable impedance variations in space and frequency.  This field may be further modified by the properties of the boundaries.  Dry wall on studs and suspended flooring tends to pass energy in the mid-to-high bass frequencies quite well, so the air impedance may not vary as much in the room.  Whereas, with concrete walls and floor, very little mid-to-high bass energy escapes the room leading to a very modal response with considerably higher impedance in the "peak" regions of space and considerably lower impedance in the "null" regions.

 

There may be different physical relations at work here.

I also think the acoustic coupling properties of the medium is what determines what kind of sound field properties will create vibration, or movement.

 

Clothing and skin may be more sensitive to velocity, and velocity even out of phase with pressure may work.

If there is no driving pressure, the air particle velocity will just disappear when the wave hits something with higher resistance, such as the floor.

 

Mid and upper bass impact, which works more on the solid parts of the body, would then require more intensity - there has to be pressure, as well as velocity.

A sound field with high intensity is a plane wave, in free field.

Such conditions we find outdoor, at distances > wavelength.

This hypothesis is supported by the general impression that the tactile impact is better outdoor, or in very large rooms.

 

At very low freq, in a small sealed room, the intensity is very low, because there is no transfer of energy.

I think it may be possible that the lowest frequencies, <15Hz, is mostly sensed by pressure sensations in the ear, and floor movement.

From the fig (in one of the above posts) we see the resonance for the legs are low, and since the legs are placed on the floor, this movement is easily sensed.

Pressure inside the room creates a large force, F=p*Area, which moves the floor, even when there is very little air particle velocity.

The velocity due to the floor actually moving is so small that it is not significant.

Link to comment
Share on other sites

Or the opposite: high PVL and low SPL. This is what is typically observed in the near field based on the articles I sourced in my original test (in the nearfield, PVL is greater). When in the near field, PVL and SPL are out of phase causing a wider gap (and resulting higher impedance) between the two compared to when they are in phase in the far field.

 

This also can be seen in okv's responses above.

No.  For some pressure P, particle velocity v, and area A, impedance Z = P/(A*v).  High PVL and low SPL implies low impedance.  These quantities are in ratio.  Being out of phase just means that the phase of the impedance is different than it would be otherwise.  The magnitude of Z is unchanged by the phases of P and v.

Link to comment
Share on other sites

No.  For some pressure P, particle velocity v, and area A, impedance Z = P/(A*v).  High PVL and low SPL implies low impedance.  These quantities are in ratio.  Being out of phase just means that the phase of the impedance is different than it would be otherwise.  The magnitude of Z is unchanged by the phases of P and v.

 

The acoustic impedance is not a constant, it changes with the properties of the sound wave.

For a plane wave, it is constant, z_a=c0*d.

In other situations, the z_a can change as a function of position, and it can consist of a real part and an imaginary part.

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.


×
×
  • Create New...