Jump to content
Sign in to follow this  
Kvalsvoll

Bulding the Room2 listening room

Recommended Posts

@SME, more knowledge often reveals that things are more complicated than you once believed, and you realize you don't have a simple answer anymore.

Frequency response is always the first we look at, and its destroys the day as we see it is not completely smooth, there are obvious flaws that needs to be corrected. As we learn more, gain more knowledge and experience, we realize this is just one visualization of what goes on, and reading a frequency response graph to actually get useful information out of it is not that easy. It is a steady-state visualization, and most of the signals we want to reproduce are transient in nature. 

Right now I am working on bass-systems. Trying to find universally applicable methods and rules to set up the subwoofers. Does not help with a very good subwoofer, if it is not set up and calibrated properly. And in most practical cases, room acoustics efficiently destroys the possibility to achieve perfect sound in a simple and predictable way.

Measurements are accurate and very useful as a tool to compare before and after when doing changes on the system. But to be able to say exactly how it sounds, by merely looking at the measurements, that is not easy.

When I observe something about the sound, I try to find a way to objectively measure and quantify what I hear. Then perhaps i can be better at predicting how it sounds, and find ways to improve faults. The first thing, though, is to verify what I think I hear - not as easy as one might believe. Often it is necessary to rig experiments.

Share this post


Link to post
Share on other sites

Just to continue, in between measuring the V110 in detail..

Frequency response is useful, as a tool - when used right. Tonal balance, potential resonances, getting the relative levels correct. It is also good for presenting a nice, smooth graph, after some heavy smoothing. 

Impulse is nice to look at, I use step response for bass. But it doesn't tell much about the sound. There is little correlation between perceived sound and the shape of the curve.

Decay is important. Here the peaks that shows in freq can be analyzed, if they are resonances those will stand out as ridges, easy to see.

Waterfall is great for show-off.

Spectrogram is the other important graph. It shows how the sound start and stop, and reveals problematic reflections and decay.

Share this post


Link to post
Share on other sites

How can you set up a bass-system if you have no measurement rig. Believe it or not, most people actually does not have acoustic measurement equipment.

Here is how:

1. Download and install a spectrum analyzer app on mobile.

2. Download the <don't remember the name> full frequency range spectrum pink noise file from my web site.

3. Play the file on repeat and monitor what happens to the sound using the spectrum app.

 

What can be fixed using this approach:

- Setting level for bass system.

- Adjust delay for main speakers.

- Add custom parametric filters for horrible and obvious resonances.

 

Very far from REW or similar, but compared to going totally blind this can fix integration and calibration to a level where many will be happy with the results.

Share this post


Link to post
Share on other sites

A big caveat here is that the response of phone mics is a big unknown.

I have a Google Pixel XL, and I recently installed an SPL meter / analyzer app.  I figured it'd be helpful for a quick SPL measurement while playing music, without having to get up and go get my SPL meter.  The app is pretty comprehensive and allows selection of Z-weighting for flat response, in the app at least.  Unfortunately, the mic on my phone seems to have much diminished response in the sub range and also has poor dynamic range.  IIRC, it taps out in the 80s dB, which is pretty much useless for measuring music that's loud enough to be fun.  :)

The sub response  on the phone seems to be bad enough that it'd be useless for doing sub measurements.  I wonder if they high pass it on purpose to improve speech clarity and reduce unwanted clipping?  Oh well.

Share this post


Link to post
Share on other sites

@SME, I only tested with a HTC, and that one looks good below around 1-2K, and down to at least 10hz.

I have observed the SPL meter apps are popular, and they look nice and advanced, but I have always had my doubts regarding accuracy of those. Even a decent measurement mic needs to be calibrated, so what can you expect from a very cheap mic inside a mobile..

Dynamic range should be a concern for spl meter app, but for pink noise freq response can be assumed that the level is kept fairly low. I noticed there was not much correlation above around 5-8k, but below say 1k the response matched REW very well.

Those spectrum apps must be set up properly, to show a graph with stable and correct response. This complicates the use. If I write my own app, it can be tailored specifically for this purpose, no settings or tweaking needed.

If many phones have significant deviation from flat below 1K, that is actually a show-stopper for any app for speaker calibration. There is a point here were total complexity and cost becomes higher than the simplest REW + mic system - which actually works.

Share this post


Link to post
Share on other sites

Posted this eminent solution on FB. It is basically useless for its intended purpose, but some will find it cool to use the phone to calibrate the stereo.

Some phones will work, some will not. There are no speccs available, so you just need to try a phone and see what happens. If you have my horns, you can place the phone in the horn mouth, and see if you get a reasonably flat graph down to below 20hz - if you don't, the phone is useless.

28419222_606958666309756_602220575004924

Share this post


Link to post
Share on other sites

Another audio-myth busted: Vibration damping platforms for loudspeakers.

Place your subwoofers on an isolation platform, and reduce vibration and noise emitted though floor to keep your neighbor happy.

But this does not work. Because it is the acoustic energy in the room that excites walls and floor to create vibration, it is not due to the subwoofer jumping up and down. Actually, most subwoofers remain quite stable, even at excessive loudness levels, and they do not transfer much motion down to the floor.

We know that - either from experience or simply because we have some basic understanding of acoustics and physics involved.

I decided to measure this, to verify whether it works, or is it one more example where basic theoretical principles and engineering trumps the marketing department.

The article is up on the web-site in the blog section, but only in Norwegian language. I will post some links to measurement results, and make some comments here. 

One V110 was measured, both spl at listening position, and floor vibration, for 2 cases - original solid base, and blocks of foam creating a completely decoupled and dangerously unstable base.

Results reveal that there is no reduction in vibration level with the decoupling - as expected. There is also no significant change in frequency or time domain at the listening position.

SPL difference at lp:

spl-lp-diff.png

Vibration difference at floor below lp:

vibrasjon-diff.png

We see that the vibration response did change, but not enough to be noticeable, and at very low frequencies there is no change at all.

The decay plots also show very little difference. However, at higher frequencies, above 100Hz, there may be a difference, depends on the loudspeaker and the floor. A good isolator can have a positive effect on vibration and decay higher up in frequency, and thus impact sound quality. But on the V110 the effect is so small it is very unlikely to make any audible difference.

On a different floor, such as concrete in a basement, the situation will be very different, with no movement at all at very low frequencies. 

Share this post


Link to post
Share on other sites

I agree with you that subwoofers induce vibrations in room surfaces via transmission of sound through the air; however, most subwoofers also transmit substantial vibrations directly into the floor.  Just because the subwoofer is not visibly jumping up and down does not mean that it is not transmitting oscillating forces into the floor.  My experience transitioning to dual-opposed subs suggests that this contribution may be very substantial.

The problem with devices that claim to decouple subwoofer vibrations from the floor is that they don't actually work as claimed.  I wouldn't put any trust in a device that does not come with engineering specs.  Such specs should indicate in a table or plot: percentage isolation (or "tan-delta" or "loss coefficient") vs. frequency vs. mass applied.  A good isolation system will isolate the object well for frequencies well above the system natural frequency and relatively poorly at and below the natural frequency.  The natural frequency depends on the mass of the object to be isolated.  I'm not aware of any isolation product targeted to the audio market that includes this information.  Therefore, I doubt that any isolation products marketed for audio actually do what they claim.

Good isolation systems typically follow one of two different approaches.  For lower mass objects, the best isolators are made using viscoelastic polymers.  Rubber is often described as viscoelastic, but this is not really accurate.  Rubber is an elastic solid but is not viscoelastic.  A viscoelastic material exhibits the viscous flow property of liquids in addition to be elastic like rubber.  The viscous properties are necessary for damping within the material to occur, which is necessary to reduce energy transmission through the isolator near the natural resonance frequency or at harmonics.

For higher mass objects (usually things more massive than subwoofers), a system using springs and/or hydraulic dampers may be used.  An example of such systems are seen in the suspension systems of cars, which help isolate high frequency (but not so much low frequency) vibrations of the road from the rest of the car.  Another example are the kinds of systems used to make buildings earthquake safe by decoupling the buildings from the high intensity ULF vibrations that occur during such events.

When using viscoelastic polymers for isolation, both the material and geometry must be optimized for the application, which depends on the mass and shape of the object to be isolated and how low in frequency the isolation should work.  The material itself must be optimized to provide sufficient rigidity but retain sufficient damping capability within the temperature range of interest.  The natural frequency of the elastic system depends on the inherent stiffness of the material (technically the Young's modulus and shear modulus), which in viscoelastic materials is frequency dependent.  It also depends on the geometry of the isolator.  The geometry is very important because it affects how the weight is distributed throughout the material and how the material deforms around it.  A large slab of material, as is typically found in commercial products spreads the mass over a very wide area, and therefore tends to have a very high natural resonance frequency.  This means that such products don't really isolate low frequencies at all.

An effective viscoelastic low frequency isolation system will typically consist of one or (usually) more isolators that have a geometry that's relatively tall and narrow but not so much as to be prone to buckling or other kinds of mechanical failure.  With proper design and optimal material choice, it is possible to provide good isolation for objects to as low as ~10 Hz, which is quite useful for decoupling subwoofers from a floor.

One viscoelastic material that is relatively easy to obtain but is rather expensive is called Sorbothane.  It is an engineered polyurethane material available in several different hardnesses.  The manufacturer publishes engineering guides from which one can estimate the properties of a system in order to choose the optimal hardness and geometric design.  I actually use small squares of the stuff under my speakers to get ~ > 30 Hz isolation, and it works.  The material can be bought as sheets, strips, feet, and other forms for different applications.  While other materials exist with similar damping capabilities from specialty manufacturers like 3M, I'm not aware of how to obtain them easily without purchasing them in extremely large quantities.

Anyway, sorry for the long post, but I just want to clarify the nature of the myth.  Subwoofers do transmit vibrations directly to the floor.  The myth is that the isolation "slabs" sold for speakers and subwoofers actually do something, when they don't (at least for low frequencies).  Nevertheless, isolation *is* possible with a properly engineered system.  It can get expensive though.  Often a better option is to build a dual-opposed sub design.

Keep in mind too that vibration isolation may also be useful for video projectors installed in rooms with heavily subwoofer induced vibrations.

Share this post


Link to post
Share on other sites
12 hours ago, SME said:

...

Subwoofers do transmit vibrations directly to the floor.  The myth is that the isolation "slabs" sold for speakers and subwoofers actually do something, when they don't (at least for low frequencies).  Nevertheless, isolation *is* possible with a properly engineered system.  It can get expensive though.  Often a better option is to build a dual-opposed sub design.

The measurements and the article explains how this works, and why it is so.

It is explained and verified by measurements that vibrations and movement in the floor is caused by acoustic energy in the room, and that any directly induced vibration by mechanical transfer from the subwoofer to the floor is not significant.

The decoupling requires no advanced science, anything that gives a resonance frequency below around 5hz will work. The foam I used is very suitable for this purpose. 

Commercially available isolation platforms have specifications for resonance frequency. Obviously the platform must be dimensioned to match the weight and surface area of the subwoofer. The flaw here is not the claimed function of the product - it will isolate the subwoofer mechanically from the floor. The flaw is the assumption that this will give less vibration in the floor, which it does not, because it is not the mechanical vibration of the subwoofer that causes the floor to move.

I may - or may not - do an English language translation of the article. Fixed it, here it is:

https://www.kvalsvoll.com/blog/2018/03/29/myth-or-fact-vibration-damping-platforms-for-loudspeakers/

Share this post


Link to post
Share on other sites
On 3/30/2018 at 10:09 AM, Kvalsvoll said:

The measurements and the article explains how this works, and why it is so.

It is explained and verified by measurements that vibrations and movement in the floor is caused by acoustic energy in the room, and that any directly induced vibration by mechanical transfer from the subwoofer to the floor is not significant.

The decoupling requires no advanced science, anything that gives a resonance frequency below around 5hz will work. The foam I used is very suitable for this purpose.

The assumption in bold is stated above and in the article as *unquestionable fact* without any evidence or theoretical physical analysis to justify it.  The rest of the conclusions are based on that unverified assumption being correct.  However, my argument is that foam products do not provide good low frequency isolation, and I stand by that argument.

On 3/30/2018 at 10:09 AM, Kvalsvoll said:

Commercially available isolation platforms have specifications for resonance frequency. Obviously the platform must be dimensioned to match the weight and surface area of the subwoofer. The flaw here is not the claimed function of the product - it will isolate the subwoofer mechanically from the floor. The flaw is the assumption that this will give less vibration in the floor, which it does not, because it is not the mechanical vibration of the subwoofer that causes the floor to move.

Do they?  I have never seen such a specification published for any isolation product targeted to the audio market.  For example, the spec sheet for the popular Aurelex Subdude HT-II platform claims to "decouple your subwoofer from the floor" but does not give any quantitative performance specifications whatsoever.  No platform is capable of providing total isolation at all frequencies, so without quantitative information, it is pointless to try to evaluate their claims by looking at the specs.

Now, if you want to see a legitimate example of a vibration isolation solution, checkout the offerings from minus k technology.  The embedded video is a very cool depiction of what their tech is capable of.  Also read their application list where you'll find many examples of applications where working vibration isolation is critically important.  Note that audio does not appear on the list, which is not to say that audio is not necessarily a useful application.  It's just that unlike audio in which products that don't work still manage to sell well because of placebo effects and a misunderstanding of physics ,the applications listed depend on tech *that actually works* with consequences that are obvious if it doesn't work.

Note that their product also includes proper engineering test data compared to another legitimate but lower performing competitor's product:

Transmissibility Curve

Read the fine-print concerning the test procedures, and you will learn that the test platform is supporting a weight of 650 lbs.

Note that I'm not trying to endorse the above product over others that may be available.  I'm merely giving an example of a legitimate vibration isolation product, to contrast it with the fake products being peddled by multiple vendors to the audio marketplace.  It's not even a matter of advanced vs. basic science.  It's a matter of science vs. pseudoscience.  Likewise, without data or at least some theoretical justification, it's anyone's guess how well any particular product (including the foam you are using for your tests) performs in reality.  As I've said many time, foam is a poor low frequency isolator, and in all likelihood, the foam used in your tests provides near zero isolation at the low frequencies of interest.

******

I will add that I have directly witnessed evidence of subwoofers transmitting vibrations directly into a floor.  I already mentioned my experience going from front/side-firing single driver subs to D.O. subs, but I observed much more dramatic effects when free-air testing my new 21" UH-21v1 drivers.  For the test, I placed a single driver face up on my living room floor and fed it sine waves at varying amplitude and frequency.  In free air, the driver produced very little actual sound below 60 Hz or so, but vibration levels rose rapidly below, down to an apparent peak at 30 Hz.  At 30 Hz, the vibration was alarmingly strong with even a tiny amount of input power (i.e. < 1 W).  The vibration was transmitted efficiently throughout my the structure of my house.  I could hear my attic vents rattling, which means that neighbors could probably hear the noise if they were outside.  If I fed that sub more like 1000W to take it to Xmax, I fear I would have caused actual structural damage!  In contrast, after installing the driver in cabinets in a D.O. configuration, there is almost no perceivable vibration at 30 Hz, even at very high levels.  At such high levels, the sound still creates a sensation, but it is that strong "back massage" sensation with no shaking whatsoever.  It is a completely night-and-day difference.

******

Lastly, let me support my argument for direct vibration transmission further using an analysis based on actual physics.  The force that a subwoofer motor imparts on the cone is always matched by an opposing force on the "stationary" parts of the driver, which is also transmitted into the cabinet that the driver is mounted in.  In a D.O. configuration, each driver contributes opposing forces on the cabinet which cancel, so there is no force imparted to the floor.  Otherwise the cabinet itself experiences a net force from the driver(s), which must be canceled by a force from the floor in order for the cabinet to remain stationary relative to the floor.  So if the cabinet is not D.O. and is well anchored to the floor (not hopping or walking), the cabinet *must* be transmitting forces into the floor.  The direction of these forces depends on the orientation of the driver(s).  For up-down firing, the force will act perpendicular to the floor.  For front-firing, the force will act parallel to the floor.  Either way, how the floor reacts depends entirely on the properties of the construction and crucially, the magnitude of the forces involved.

So let's talk about these forces using real quantities.  For brevity, I will give a formula that relates the force between the moving assembly and the rest of the driver (let's call it the "recoil force") in terms of SPL at 2 meter ground-plane for a sealed cabinet operating under far-field conditions.  For a single driver sealed system, this is the force that will be transmitted to the cabinet and to the floor.  This can be readily related to measurements published here on DataBass.  Upon request, I'm more than happy to show a detailed derivation of the formula below.  It is derived using "Newton's first law of motion", the relationship between driver acceleration and SPL at a distance, and the definition of SPL in terms of pressure:

    Recoil Force = (0.000411 m^3/s^2) * 10^(SPL_@_2m_GP / 20) * Mms / Sd

For unit consistency, Mms must be specified as kg and Sd as m^2.  As general points of interest note that for a given SPL, higher Mms increases recoil force, and higher Sd decreases recoil force.  Let's throw in some typical numbers and see what happens.  For example, we'll consider output of 120 dB SPL RMS at 2 meter GP from a "typical" deep-bass 18" driver with an Mms of 0.75 kg, and Sd of 0.120 m^2:

    Recoil Force = 2570 N = 578 lbf

These are RMS numbers.  The respective peak quantities are 3630 N and 817 lb.  For a down-firing configuration then, imagine a 1634 lb (740 kg) weight (doubled because of peak-to-peak difference in force) being repeatedly added and removed to the location of the sub and you get the idea of how strong the forces can be.  This is another reason why stout cabinet construction is important, not just to maximize efficiency of the driver(s) but to also maintain cabinet integrity at high output levels.  Of course, 120 dB SPL @ 2 meter GP is running things pretty hot (max capability for a lot of 18" drivers), but the forces involved are still significant at lower levels.  For example, at 100 dB SPL @ 2 meter, 20 dB less, the forces will be 1/10th as much, which is likely to still be enough to vibrate a suspended floor.

******

So to summarize, direct vibration transmission between a cabinet and floor can be quite substantial and almost certainly contributes noticeable vibration in addition to the vibration induced by the sound itself.  This transmission occurs even if the subwoofer appears relatively motionless.  If the subwoofer is actually visibly moving, the low frequency vibration transmission may be reduced, but the repeated collision between the sub and floor will create a lot of unpleasant high frequency rattling noise.  And of course, a moving subwoofer is potentially unstable if it's part of a stack.  Reading other threads here, I know @Ricci has reported that "sub walking" can be a serious issue in a pro setting where they are often pushed to peak levels for an extended period of time.

The real reason that sub isolation platforms fail to reduce vibration is that they don't work as advertised.  While foam can prevent the unpleasant high frequency noise that occurs when a vibrating sub or speaker rattles against a floor, it does little to isolate the low frequency vibrations of the sub from being transmitted directly to the floor.  An apparent lack of visible motion is useless as an indication of isolation.  In fact, an isolated subwoofer will move *more* than one that is not isolated.  That's because in order for the forces transmitted to the floor to be zero, a net force must be acting on the cabinet, which will cause it to vibrate in-place instead.

Share this post


Link to post
Share on other sites
7 hours ago, SME said:

The assumption in bold is stated above and in the article as *unquestionable fact* without any evidence or theoretical physical analysis to justify it.  The rest of the conclusions are based on that unverified assumption being correct.  However, my argument is that foam products do not provide good low frequency isolation, and I stand by that argument.

Do they?  I have never seen such a specification published for any isolation product targeted to the audio market.  For example, the spec sheet for the popular Aurelex Subdude HT-II platform claims to "decouple your subwoofer from the floor" but does not give any quantitative performance specifications whatsoever.  No platform is capable of providing total isolation at all frequencies, so without quantitative information, it is pointless to try to evaluate their claims by looking at the specs.

Focusing on only one part makes it easier to follow:

The decoupling I used works. I knew before I did the measurements, actually I did the measurements because I had a readily available decoupling solution that works TECHNICALLY. However, this solution is not suitable for ordinary use due to looks and the fact that the subwoofer floats on top and is very unstable. I had the blocks because i use them to handle the subwoofers when moving them around.

Transfer of vibrations are determined by only 2 parameters - resonance frequency and damping. If the resonance frequency is low enough, and the damping is not too high, it works.

The problem with foam is that the spring stiffness increases in a nonlinear fashion when compressed. If the foam is too soft and too low, it will compress and give a resonance frequency that is too high. The product you link to here (spec-sheet) can not work, because there is only 1 inch foam to act as the spring. this product may very well work to remove vibrations at higher frequencies, but it will not decouple at very low frequencies. If I had used a similar product or solution, it would be correct to assume that the measurements are not valid becuase the subwoofer is in fact not decoupled at low frequencies.

Share this post


Link to post
Share on other sites
12 hours ago, Kvalsvoll said:

Focusing on only one part makes it easier to follow:

The decoupling I used works. I knew before I did the measurements, actually I did the measurements because I had a readily available decoupling solution that works TECHNICALLY. However, this solution is not suitable for ordinary use due to looks and the fact that the subwoofer floats on top and is very unstable. I had the blocks because i use them to handle the subwoofers when moving them around.

Transfer of vibrations are determined by only 2 parameters - resonance frequency and damping. If the resonance frequency is low enough, and the damping is not too high, it works.

The problem with foam is that the spring stiffness increases in a nonlinear fashion when compressed. If the foam is too soft and too low, it will compress and give a resonance frequency that is too high. The product you link to here (spec-sheet) can not work, because there is only 1 inch foam to act as the spring. this product may very well work to remove vibrations at higher frequencies, but it will not decouple at very low frequencies. If I had used a similar product or solution, it would be correct to assume that the measurements are not valid becuase the subwoofer is in fact not decoupled at low frequencies.

The Auralex products are probably the most popular for subwoofer and speaker "isolation".  Most products I've seen marketed for audio use are very similar.  At least we agree that such products are not capable of working as advertised.

I'm still skeptical about the foam blocks you describe, but without data I can't know for sure either way.  And even if they are providing isolation, it's possible that your floor is just more resilient to direct transmission.  Perhaps the floor vibration you do measure comes mostly via the walls, which may be made to vibrate by the sound alone.

I do think there is potential for a sub isolation system, one that actually works :P, to reduce unwanted vibration and possibly reduce neighbor annoyance.  However, the approach will be hit or miss.  It's going to depend on the details of the construction and how vibrations induced directly vs. via sound are transmitted throughout the structure.  And because an effective isolation system is likely to be impractical, expensive, or both, it makes much more sense to opt for D.O. subs if possible.  I did that, and I unquestionably (albeit by subjective opinions of myself and others who experienced before vs. after) experienced a substantial reduction in structural vibration, even though it was not eliminated completely.

Share this post


Link to post
Share on other sites
11 hours ago, SME said:

The Auralex products are probably the most popular for subwoofer and speaker "isolation".  Most products I've seen marketed for audio use are very similar.  At least we agree that such products are not capable of working as advertised.

I'm still skeptical about the foam blocks you describe, but without data I can't know for sure either way.  And even if they are providing isolation, it's possible that your floor is just more resilient to direct transmission.  Perhaps the floor vibration you do measure comes mostly via the walls, which may be made to vibrate by the sound alone.

I do think there is potential for a sub isolation system, one that actually works :P, to reduce unwanted vibration and possibly reduce neighbor annoyance.  However, the approach will be hit or miss.  It's going to depend on the details of the construction and how vibrations induced directly vs. via sound are transmitted throughout the structure.  And because an effective isolation system is likely to be impractical, expensive, or both, it makes much more sense to opt for D.O. subs if possible.  I did that, and I unquestionably (albeit by subjective opinions of myself and others who experienced before vs. after) experienced a substantial reduction in structural vibration, even though it was not eliminated completely.

After looking into specifications - which I could not find, because there are none - for the product you mention, I now looked into a different product, where I was informed that the resonance frequency was 5hz. I could not find any information on the product's web site to verify this claim. I could not find any relevant technical information. So I stand corrected about the specifications for those products, and the description reads like snake-oil to me.

If the spring is too hard, it will place the resonance right in the active working range of the subwoofer, and potentially make things worse. It also need to decouple in all directions - pivot, fore-aft, up-down, rotational.

It is easy to verify if it can work. If the spring is compressed less than several inches/cm, it is too stiff. If you measure the deflection of the spring, the resonance frequency can be calculated, provided the spring is linear. A foam pad is not linear.

If you make one large foam platform , sufficiently thick and soft to provide a low resonance, it would still be too stiff on the fore-aft direction, because the supported area is too large. A platform with springs in the corners could work. The foam blocks I used are only suitable for experiments, the subwoofer is not stable on top.

But the most important observation here is that it makes no difference, even if the decoupling is made so that it significantly reduces mechanical coupling down to frequencies below working range.

The floor in the room used for measuring this is what I consider quite normal for a wooden floor. But the subwoofer is placed close to walls, close to a corner. And in that location, the floor is much more rigid than in the center of the room, because the beams supporting the floorboards are supported only at the ends where the floor meets the walls.

The subwoofer vibrates, it moves, and it is transmitting those vibrations to the surface it sits on. It just that the level is too low to be significant, compared to the effect from the acoustic sound pressure in the room, which acts on the whole surface of all walls and ceiling and floor.

And the reports from people who tried this, says the same - no sound reduction for the neighbor.

The V110 is different from the usual sealed box subwoofer with long-excursion, heavy-moving-mass driver. The moving mechanical mass is 118g - very low compared to the usual around 500g or more. But the driving force is similar or higher. And it is quite tall, creates a large momentum. The main force from the surroundings acting on the cabinet will be the acoustic load on the port exit, and this load is far less than the typical 500g mechanical mass. The 118g cone assembly does not move much, due to the acoustic loading, so even comparing the moving mass to a different subwoofer is not relevant. Still, this subwoofer certainly moves and vibrates. The older ancestor T138 in the media room is more lightweight, they move and have to be pushed back into location occasionally.

A dual-opposite design is one solution to remove low frequency vibration, that actually works. But the floor still vibrates.

Share this post


Link to post
Share on other sites

I did my force analysis for a sealed box sub and found the results interesting.  Keeping SPL constant, the force is essentially frequency independent.  I'm curious how it would change for a ported sub and/or horn sub, but those would be a bit harder.  My guess is that the ported subs would get a bit more SPL vs. force near tune, and the horn would get more SPL vs. force in its prime range.  I'm not exactly certain though.  A more complex analysis would look directly at the force balance on the moving vs. stationary assembly of the driver, which must take into account motor force, back EMF effect, and the differential pressure on each side of the cone.  For a sealed box, the latter is quite trivial to evaluate, but the situation is much more complicated for ported boxes and horns.

As for neighbor complaints, a lot depends on the circumstances.  If the house units are detached, it's likely to be relatively high frequency bass that's more offensive.  Content at around 100 Hz is relatively easy to hear at low levels and often passes very efficiently through walls and windows.  I believe some triple pane window designs can be even worse because the middle pane acts as a resonator.  I once worked in an office in which the windows had a strong resonance at around 100 Hz, which was very annoying at times.  Thunder tended to create a very loud tonal sound, and I recall one day in which a lawn mower being operated hit the resonant frequency and was annoyingly amplified.

Share this post


Link to post
Share on other sites

Most of the time direct vibration transmission from the sub enclosure, into the floor and structure, is probably negligible, or of much less concern than the vibration caused by acoustic transmission. However we can't make a blanket statement and say it is never an issue. A worst case scenario might be something like: A suspended floor,  hard surface, down firing driver/s or PR's which are not mirrored, very high mms, lightweight enclosure, sub being driven very hard. In this hypothetically bad scenario or something similar it may indeed be an issue. For most typical sub designs in a carpeted space it's unlikely to be worth addressing IMHO. 

Share this post


Link to post
Share on other sites

I think it's helpful to distinguish between direct vibration transmission that's a problem, where a sub is rattling against the floor or not staying in place, and vibration transmission that is not a problem but still contributes to tactile feel.

I was honestly very surprised by how much tactile feeling I gave up going from the old pair of single-driver ported subs to D.O. sealed subs.  Partly as a consequence, I'm running my bass quite a bit hotter than I used to.  I know I'm not alone here either.  I've seen people on AVSForum report similar observations.  Then there was that crazy guy who purposely ran his HS-24s free air for a while and reported all kinds of infrasonic shaking effects, which were probably entirely from vibration instead of SPL, and actually did damage his house.

FWIW, the ported subs were on rubber feet and at high output definitely shook a lot.  I recall once or twice putting a glass of water on a coaster on top of the Hsu VTF-15H sub while it was playing the low organ notes in the Saint-Saens Symphony and marveling at the patterns that developed on the surface of the water.  This didn't happen with the glass placed on any other room surface.  By comparison, the new subs are inert except at the few frequencies (i.e. 11 Hz and 19-24 Hz) where the floor under them shakes, and only then at higher SPL than it used to take.

It's possible that @Kvalsvoll 's testing involves a floor that's firm enough to resist most direct transmission, and the fact that he's testing with horns probably means more SPL vs. vibration from the same drivers.  Another thought is that multiple subs spread out may exhibit more or less vibration depending on how they combine in the air vs. combine in the floor system.  There are lots of variables here.

Share this post


Link to post
Share on other sites

I can assure you the floor is not more stable and firm and damped than any other typical house. It is a very old house, so the walls are heavy, but the floor does not seem to be more rigid than in typical newer houses. It gives a nice and pleasant tactile effect, from around 20hz and down towards 10-12hz.

The problem you mention SME, about resonating and vibrating coupling between any loudspeaker and the floor certainly can cause audible problems, but it is also easy to fix, even those isopod products should work well for that. I consider this to be something that should be designed in to the speaker/subwoofer, so that the customer does not find the need to buy additional products to fix rattling and noises.

When you have something capable of decent output full-range, clean, then all kinds of rattles and noises suddenly appear, often from outside the room you are listening in. On the processor in Room2 there is a small strip of tape attached to the acrylic display cover - it rattles. The spotlight assembly in the ceiling in The Moderate Cinema rattled, then there are the structural noises from the house, which can not be fixed so easily.

The measurements were done with only one V110. Really no reason to complicate things by trying to decouple several subwoofer units.

Share this post


Link to post
Share on other sites
7 hours ago, SME said:

...

I was honestly very surprised by how much tactile feeling I gave up going from the old pair of single-driver ported subs to D.O. sealed subs.  Partly as a consequence, I'm running my bass quite a bit hotter than I used to.  I know I'm not alone here either.  I've seen people on AVSForum report similar observations.  Then there was that crazy guy who purposely ran his HS-24s free air for a while and reported all kinds of infrasonic shaking effects, which were probably entirely from vibration instead of SPL, and actually did damage his house.

...

I really should find something new for another article to post, so we can continue to go far-too-deep into something else, don't think we disagree enough to continue this one much more now.. I have one, most of the text is ready, there are pictures and measurements, perhaps tomorrow. 

I think @Ricci's post sums it up quite nice - mostly, not an issue, but, depends - some subwoofers, placements, floor and building issues, can cause problems.

I still see your example here as surprising, even when considering that those subwoofers will create much more force than the very different V110. To give a substantial difference in tactile feel, there must be a very significant difference in vibration level. And this can be measured. It should show up on the frequency response, and can be measured with a mobile accelerometer app.

It is actually possible to partially replicate this experiment; you could run only one driver active and short the other one - it won't be the same as single driver as the dead driver will still move a little, but should be enough to show a difference in vibration level form the subwoofer. With 2 drivers it should be dead, with one there will be motion.

And just when writing this, I read another difference - from PORTED to sealed. That can also be a factor in tactile feel experience, but is it really that huge..

Maybe I should repeat the measurements comparing sealed and ported/horn. Difficult to seal off the horn output in a V110, and there may actually be too little output left even for a measurement only.

Share this post


Link to post
Share on other sites
1 hour ago, Kvalsvoll said:

It is actually possible to partially replicate this experiment; you could run only one driver active and short the other one - it won't be the same as single driver as the dead driver will still move a little, but should be enough to show a difference in vibration level form the subwoofer. With 2 drivers it should be dead, with one there will be motion.

Hmm, I think I could could do this pretty easily actually.  The drivers are in isolated chambers, and they are wired in parallel.  All I have to do is disconnect the SpeakOn on one side of each, I think.  The only issue is that the gap between them and the wall may not be big enough to cleanly disconnect the cables, and I'm not willing to pull them out from the wall because the speakers are on top and are precisely aligned.

Share this post


Link to post
Share on other sites
13 hours ago, SME said:

Hmm, I think I could could do this pretty easily actually.  The drivers are in isolated chambers, and they are wired in parallel.  All I have to do is disconnect the SpeakOn on one side of each, I think.  The only issue is that the gap between them and the wall may not be big enough to cleanly disconnect the cables, and I'm not willing to pull them out from the wall because the speakers are on top and are precisely aligned.

And you need some method to measure it, doesn't need to be accurate or calibrated, just to be able to compare.

Share this post


Link to post
Share on other sites

Looking for a reasonably priced power amp for midbass use - 2 channels, around 500-1000W/8 ohms/channel, NOISE-FREE.

Something like the inuke 3000, without the noise.

I have a design for some very powerful and compact midbass horns, and now I found a customer for them, so if we can find a suitable amplifier they will be built and tested. My SA-700 can be used, but it is kind of overkill, and quite expensive.

Any tips for such an amplifier? And no fan-mods or similar tweaking, it must work out-of-the-box.

Share this post


Link to post
Share on other sites

By noise free, do you mean acoustic (i.e. fans), electric( i.e. hiss), or both?

If you're open to plate amps, you might look at some of the lower power offerings from SpeakerPower.  They work best with 4 ohm, however. I believe they are based on the ICE amps, so you may be able to find other amp designs based off of ICE in EU.  (Most of the ICEs I think like 4 ohm more.)  Finding anything that's acoustic noise free with that power level will be tough.

Share this post


Link to post
Share on other sites
15 minutes ago, SME said:

By noise free, do you mean acoustic (i.e. fans), electric( i.e. hiss), or both?

If you're open to plate amps, you might look at some of the lower power offerings from SpeakerPower.  They work best with 4 ohm, however. I believe they are based on the ICE amps, so you may be able to find other amp designs based off of ICE in EU.  (Most of the ICEs I think like 4 ohm more.)  Finding anything that's acoustic noise free with that power level will be tough.

Acoustic noise - fan noise.

Signal noise is rarely a problem with power amplifiers.

Plate amp is a no-go, requires cabinet building. They also tend to be more expensive.

Crown could work, I see they have the xli 2500 with regulated fan. Or if we could find a used K2.

Share this post


Link to post
Share on other sites

Sound field properties are important for how low frequencies are perceived.

Sound field properties are the relations between pressure, particle velocity and intensity. Both level and phase of velocity relative to pressure can change significantly inside a small room when properties of the bass system are changed - position of sound sources, delay, eq settings.

In Room2 I experiment with different bass system configurations, to find methods for setup that are repeatable and consistent. I experience differences even if the measured frequency response is the same. Some of those differences are caused by time-related issues like phase and decay. But some effects, notably at lower frequencies, are caused by differences in sound field properties. 

The system I test now is a FL FR + BL config - one additional unit located at the back of the room. In this setup there is and increase in velocity and sound power around and below 20hz, which causes more movement in the floor than desired, on many music recordings there is simply way too much ulf.

I noticed this instantly, before doing sound field measurements - too much ulf. And indeed it looks like the BL back unit changes the sound field from 20hz and down, there is more velocity in the vertical direction. 

The way to fix it is to adjust the frequency response - reduce the level around 20hz. It gets much better, but still not perceived as neutral - there is too much ulf, and noise and unintentional ulf sounds become distracting.

Also further up in frequency, there seems to be a clear correlation between velocity level and perceived bass quality - smooth is more neutral, more velocity is better and sounds more powerful and natural. Simply more fun.

Above the 40-50hz range the situation changes, velocity is still important, but will be more of a measure of sound directivity. In the upper bass range phase and timing and frequency response is the important properties to focus on, as the velocity and intensity more or less follows due to the sound waves being smaller compared to room dimensions and distance to sound sources.

 

Share this post


Link to post
Share on other sites

Very interesting observations.  How does the  FL/FR/BL config do higher in frequency velocity/phase/FR wise?

What size driver/sensor do you use for velocity measurements?   I need to actually do these measurements at some point.

JSS

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
Sign in to follow this  

×