What about Double Bass Arrays??

[807Recordings]

With all this interest in sound acoustics and such what do people think of double bass arrays?

I ran across this forum but it does not seem to have many pro's talking.

How about draw backs?

Double Bass Array (DBA) - The modern bass concept! - AVS Forum

[807Recordings]

As I read through this the most interesting parts I came across at least in terms of bass is that is seems to work best with no bass trapping.

Better transient responses,
far more accurate bass,
etc.

Where I am wondering is if this is really true then what happens when you take into other parts of acoustics like mid frequencies above the bass range etc.

[SAC]

OK, you asked...

As they are not located within 1/4 wavelength of their highest frequency reproduced, you have an EXCELLENT comb filter and polar lobing generator! Every frequency above where every combination of interdriver separation will be greater than 1/4 wavelength will simply rip your response to shreds - both in terms of frequency domain comb filtering and with regards to the concomitant spatial polar lobing!!!

If the drivers are labeled:
12
34

You will have the inter-driver spacing issues with all of the various separations for each combination such as 1-2, 1-3, and 1-4. Additionally, you must also consider the virtual sources due to boundary reflections and their spacings as well!
This violates the fundamental notion of close packed coupled drivers. Heck, with such distributed spacing, the response will even be worse than listening to a line array in the near field!

This is a classic example of what often happens when one likes the sound of one speaker, and seeks to increase the gain by adding another, and in this case, another and another. Unfortunately, it is not so easy. As in order to avoid the destructive summation (superposition) resulting in comb filtering and polar lobing errors, the drivers must either be close packed such that their acoustic centers are within 1/4 wavelength of the reproduced frequencies, or they must be summed in a combination known as a Bessel array, whereby the gain is increased without the associated frequency and spatial errors.

And what happens when you do this higher in frequency where the wavelengths are still shorter??? The degree of comb filtering and polar lobing becomes even worse!

In fact, one would be hard pressed to devise a better 'distortion/anomaly' generator! What about drawbacks? Its almost ALL drawbacks, with additional gain to add insult to injury! 1 step forward and 3 steps back. This is not the way to productively increase LF output in a quality manner.

This is a case of a classic case where the ignore-ance of fundamental acoustical principles and behavior with respect to time regarding the superposition (summation) of common signals shifted in phase comes back to bite you. The fact is that this behavior happens 'by definition'! There is not much to debate as this is not a case of speculation.

This is a significant (and an unfortunately all too commonly committed) acoustical faux pas...

Below find a diagram documenting the Comb Filtering and Polar Lobing behavior of 2 Spaced Sources:

[jdg]

i'd like to see a measurement far finer then 1/3 octave.

[807Recordings]

SAC thanks for the info.
I knew there was obvious stuff I just never looked at, such as comb filtering, etc.

Of course I do not claim to know much on this subject but I do like to see different perspectives to a problem, and why or why they don't work.

[SaSi_SiDi]

But is comb filtering really going to be an issue with bass arrays? The problem of comb filtering becomes a problem when the separation becomes comparable to the frequencies reproduced.

At 80Hz, the wavelength is 4.25m and a subwoofer is expected to be fed with frequencies well below 100Hz.

I remember reading a Harman Kardon paper on subwoofer positioning. The bottom line result was that the more subwoofers you add in a room, the better overall result you get. It then becomes an issue of correct placement in order to get the best out of the least number of units.

[SAC]

Yes, the problem exists at low frequencies just as it does with higher frequencies. And it has a pronounced impact on the localization of the sources.

And Floyd Toole's research was simply an attempt to moderate the modal behavior over a space.
If it didn't matter, his research for a given goal would have been superfluous as, if it were not valid, it wouldn't matter!

The lower the frequency/longer the wavelength simply means the comb filtering spacing is greater than then the high er frequencies and the polar lobes are larger and fewer in number. The physics is the same.

Much has occurred in the understanding of the arraying subwoofers.

BTW, to approach it from the opposite perspective, this same destructive interference can be used to our benefit in such design techniques as a 'cardioid sub' - to provide a more controlled directional LF propagation.

And measurements confirm them.

Take a look at this model generated in EASE. It is valid for ALL sources based on the separation relative to wavelength.

[tINY]

But, your typical small room benefits greatly from multiple sub-woofers.

Assuming that you are listening somewhere near the center of the room and you have a suffciently steep LPF, then lobing isn't an issue. By placing 2 or 4 subs in the right places, all you really do is cancel/not excite room modes.

Consider a room with 4 subs in the corners half-way between floor and ceiling:

In a square room 30 feet on a side, if you stay 4 feet from the wall, the worst path-length difference is 36.8 feet. This gives you a first cancellation at 15 Hz and another at 45Hz. This could be bad, except that the other 2 woofers are 26.3 feet away and you have all of your reflected waves arriving about 1 cycle later (at various times).

Do you want to argue about bass transients? The modal ringing is already dealt with, so we are talking about the first 2-4 cycles...



-tINY

[SAC]

You mention the first cancellation -

Don't stop there! They repeat at EVERY multiple thereafter.

And the spatial polar lobing is real as well.

And the additional delay from the other sources simply exacerbate that! (e.g. another spaced null repeated at integer based intervals and concomitant polar lobing!). And yes, the reflections off the boundaries (for all frequencies greater than 1/4 wavelength) will serve as yet additional virtual sources that will combine and create more CF and PL.

Superposition is fundamental, there is no case where this is avoided. And only the Bessel array allows for the combination in such a manner to occur where the original polar and frequency response is maintained.

We use this very feature (superposition and gain of spaced drivers) as the foundation for designing 'directional' cardioid subs!

[tINY]

And how well do those subs work in 300-900 square foot rooms?

A single sub in a room that size is going to have the same problems you describe - only it's worse because the axial modes get excited and the initial delay is longer (the wave has to reflect off the back wall).

Correct me if I am wrong here, but I assume that the original post was about some room smaller than an arena or a concert hall.



-tINY

[SAC]

Superposition is superposition. It is not room size dependent! (Except that a room with more closely spaced boundaries will serve as additional virtual signal sources further compounding the issue!)

A single sub will exhibit such behavior to the degree that its placement is further than 1/4 wavelength from a boundary. Wavelengths shorter than 1/4 wavelength will effectively sum (and hence the attributes of SBIR), while those greater than 1/4 wavelength will exhibit reflections from the boundary (acting as a virtual source spaced 2X as far as the speaker is located from the boundary) that will superpose (combine/sum) destructively.

Additionally you will have room mode issues due to the pressure loading of the room with wavelengths smaller in size than the room dimensions resulting in room dependent standing waves..

Check out:
http://www.synaudcon.com/website08/V...Boundaries.pdf

Note this same report is also featured as "Hot" on Tom Danley's website as well.
Danley Sound Labs - Home of Tom Danley - Innovative Designer of Pro Loudspeakers & Subwoofers

[tINY]

Have you ever studied small room acousitcs, the internal acoustics of speaker cabinets or even reverberation in larger rooms?

There are always constuctive and destructive interference patterns - the questions are: When do they matter? And: What can we do about it?

Take some measurements. Put your cardiod sub in your living room and measure the directionality....



-tINY

[SAC]

Oh, we are supposed to have an idea about that which we speak???? Go figure!

Yeah, I spent a few hours studying at Best Buys and the 19 year old salesman explained it ALL to me! And they say they are experts! Gee, and I actually heard a sub once - a friend has one hooked to his computer.

And don't let the measurements in a small room confound you!

Amazing how physics hold sway everywhere but here on the forum (and in the world of high-end audiophoolia).

LOL!

[tINY]

Physics is a toolkit. If all you have is two-source interaction model or a 2, 1 and 1/2 Pi radiation model, you have a limited toolkit. (There is an old saw about having a hammer and everything looking like a nail...). What you really need below about 300Hz in most rooms is a transfer function that incorporates all sources and boundaries (and these tend to work over a reasonably large area).

Unless you plan on dropping yourself and a speaker system from an airplane and listening on the way down, you have to live with boundary interactions. Since most people don't have the space and money to have a large hall or anechoic chamber to listen in, practical bass reaponse in most rooms doesn't look like radiation into 1/2 Pi space....




-tINY

[SAC]

You're right! you not only have the room modal response, but you also have the comb filtering and polar issues that 'come with' comb filtering - as was measured by Pat Brown in a small bounded room and featured on Tom Danley's web site.

And while its nice to mention non sequitur items such as large halls and anechoic chambers, they have NOTHING to do with the measurements or the concept.

But its fascinating to see that some think that superposition is somehow predicated upon what space the signals are in. The fact is, the more spaced sources you have, the greater the destructive interference.

Damned that physics... and yes, it is a toolkit that has meaning - one that is not subject to taking affect only when we choose.

You haven't by chance ever been employed by Bose, have you?

[tINY]

Nah, not only do I have philisophical issues with Bose - they are in Boston... I like the west much better, thank you.

For everyone else following along - interference can be construtive or destructive. So, it you balance things right, the net effect is similar to a single sub in the middle of a football field (with some gain) ...at least in a frequency response sense.

Pat Brown's article that you refer to is pretty good - and you can draw a lot of conclusions for how to set up a dance hall with high ceilings or a concert hall from it. But, like most clasical acoustic science, there are a lot of issues in small rooms that are a bit hard to imagine without vastly more complicatied modeling.

Luckly, Floyd Toole has done some work on this subject and some fairly easy to understand presentations are available in several places.




-tINY

[SAC]

Originally Posted by tINY

Pat Brown's article that you refer to is pretty good - and you can draw a lot of conclusions for how to set up a dance hall with high ceilings or a concert hall from it. But, like most clasical acoustic science, there are a lot of issues in small rooms that are a bit hard to imagine without vastly more complicatied modeling.

You keep talking about "dance halls with high ceilings", "concert halls" and "large" rooms.
Did you read what Pat said, and what was measured in a SMALL room?

None of the behaviors mentioned have anything to do with room size, with the exception of modal behavior!!!

Comb filtering and polar lobing has no room dependency - with spaced sources and superposition - it occurs. And room modes are a small room phenomena, both of which are addressed.

And while Floyd Toole has done much in his attempts to mitigate modal behavior to establish an arrangement with the least modal variation, it has little to do with spaced sources on a vertical boundary in a small to moderate sized room - but of which ALL are "small acoustical spaces" as defined by Schroeder. Other descriptions of space size may have psychological meaning, but this additional distinctions are meaningless in acoustics.

So you have modal behavior, and you have comb filtering and polar lobing.

And repeated references to "dance halls", 'large rooms' or "concert halls" is non sequitur. And to the degree that any of them qualify as a large acoustical space as defined by Schroeder, we can then cease to care about modal behavior and are still left with the comb filtering and polar lobing anomalies!

Thus, NONE of the fundamental issues, except modal behavior, are alleviated, as modal behavior is the only aspect that is dependent upon room volume.

And ironically, only Bose has maintained (incorrectly - as was verified in the infamous SynAudCon trials at the Intelligibilty Workshop with EV, JBL and Bose in 9.1986 hosted by Rollins Brook, Dave Klepper and Victor Peutz along with participants such as Dick Heyser, Ferrel Becker, Jim Borwn, Art Noxon, Russ O’Toole, John Prohs, Larry Shank, Dave Andrews, Don Eger, Doug Jones, Hellmoth Kolbe, Peter Mapp, George Augspurger, Jim Carey, Mary Gruszka, Bruce Howse, Don Keele, Jay Mitchell, John Murray,...all of whom are names in acoustics that should be readily recognized!!! ) that speaker Q and controlled dispersion are not important aspects of a speaker's performance for the precise reasons I have stated! You might note that only Bose maintained as you have that such destructive interference would not be a problem - a problem soundly verified.

The point I am citing is not one requiring verification, as it has been very well documented and measured in the past by some of the absolute best in the business!. But I guess if we have time, we could also debate as to whether a heavier than air craft can fly. ;-))

But I am fascinated to keep hearing of some instance where spaced sources are not somehow subject to comb filtering and polar anomalies. If you indeed think you know of one (aside from the Bessel array), by all means submit and have the results published by AES/ASA! And be sure to register the technology so that you own the rights. As you will have accomplished something that has not been accomplished by anyone else to date.

[tINY]

Live Sound: Tools of the Trade: How Boundaries Affect Loudspeakers - Pro Sound Web


Are you talking about this one?

Where he states:

Overview
The test environment is a large, open space with very rigid boundaries on three surfaces.

I selected a loudspeaker line-up with some devices commonly used by contractors - a bookshelf loudspeaker, a subwoofer and a medium-format horn (Figure 2).

...and so bypasses all of the small room issues you want to ignore.

When you say

Comb filtering and polar lobing has no room dependency - with spaced sources and superposition - it occurs.

are you inferring that reflections off of room boundaries don't cause comb filtering? Or do you consider reflections off of side and back walls not to be a room dependancy? Or do you want to call it something else when considering wavelengths about 1/3 the room dimmension or longer?

Essentially, you have strong interactions between your "comb filtering and polar lobing" and your "modal behaviour" when considering several spaced LF transducers in a typical listening room.

Since we are talking about sub-woofer arrays in "small" rooms, the rest of your post is irrelevant... I'm just talking about the sub 100Hz range in room 2000-4500 CF.




-tINY

[SAC]

Have you read what has been posted????
Apparently not, as you ask things that were clearly stated in the initial posts and you keep bringing up issues that are non sequitur to the issue! And you seem to be disagreeing by positing the same room response which I asserted from the beginning - except you seem to focus on the 'room' while I was focusing on the additional problems the speaker propagation would generate...both leading to (I think) a similar conclusion. But then you keep talking about a small room as opposed to a small room - as if thee is any other space or spaces involved other than a Schroeder "small acoustical space" which determines the rules at play.

Spaced, non signal aligned transducers reproducing the same passband will interact destructively if they are further apart than 1/4 wavelength of any reproduced frequency. And while we see comb filtering in the frequency domain, spatially you have polar lobing anomalies. In other words, if the original dispersion was cardioid or even omnidirectional, you will now have a response wit the total dispersion area divided into lobes. And the higher the frequency, the more lobes you will have.

PERIOD.

That will be the nature of the direct signal propagated by the spaced sources. Add to this the room modal behavior and you will have a mess - compounded by the spaced sources. The problem with comb filtering and polar lobing could be, in large part - but not totally, mitigated by close packing or vertical close packing within 1/4 wavelength of a floor-wall boundary.

Where do you get the idea that ANYTHING bypasses small room issues! It it TOTALLY conditioned by the acoustical behavior of a Schroeder defined Small Acoustical Space! And since we are assuming reproduction of music to 30 Hz, that means any space less than about 250,000ft^3!!!

And if you understand the nature of measurements, they were taken in a "large open space" precisely to eliminate the additional reflections and comb filtering and polar anomalies introduced by additional boundaries aside from those specifically to be included in the test scenario! Rather than compromise the conclusions, this very consideration provides yet more substantiation!


And do reflections act as virtual sources???? You have to ask this? Of course they do! That concept if fundamental to the entire discussion of reflections! And yes, the spacing/time offset introduced by the real and virtual sources will introduce still additional comb filters and polar lobing in addition to the various other combination of transducer spacings!

And the distinction of the room in which measurements were taken?? It is about the same as size as the one illustrated in the example with the 4 spaced woofers mounted vertically in the walls! It is STILL a "small acoustical space" as defined by Schroeder that determines the acoustic principles at play in the space! But the significance is that rear wall reflections were not an issue! Only reflections from the near boundary surfaces. If far wall surface reflections were considered, the response would be WORSE, as you would have yet another source of superposition..

And as the measurements show, we have a combination of room modes and comb filtering and polar lobing.

And Whatever Floyd Toole wrote has NO bearing on the topology of the 4 widely spaced woofers (both from boundaries and other drivers) in a single vertical boundary! Just because he experimented with four spaced subwoofers mounted adjacent to boundaries evaluated on the horizontal plane has little to do with the topology they have built!...aside from the fact that they employed the number 4 as well. Or did Floyd also evaluate the distribution of modes on the ceiling (a VERY short wall) near the center of the room assuming a vertical boundary placement. Funny, he never mentions that.

What is fascinating is how you have created yet another acoustic space unknown to science where apparently acoustics behaves as if in some realm of the Twilight Zone.

The result in the space is a strong modal presence compounded by comb filering and spatial polar lobing.

"Essentially, you have strong interactions between your "comb filtering and polar lobing" and your "modal behavior" when considering several spaced LF transducers in a typical listening room."

Duh! You think?!?!?!? ;-)))

That has been my contention from the first post onwards! You don't simply have room modes! You have additional frequency and spatial errors as well made worse by the non-optimally spaced drivers!!!! And these frequency and spatial errors are worse than if you had just one transducer or several tightly packed drivers!

[tINY]

Originally Posted by SAC

And these frequency and spatial errors are worse than if you had just one transducer or several tightly packed drivers!

OK - I think I understand you now. Above is the only part I disagree with.

In the original drawing from Germany, the 4 woofers on the front wall are shown to be about 1/4 of the way in from the ceiling/floor and side walls. If you only consider frequencies below where the distance between the drivers is 1/2 wavelength, then most of the room sees no destructive interference from the multiple drivers (at least until the wavefront bounces off the back wall). Frequencies higher than that (where 1/2 wavelengths are less than the driver spacing) will not be coherent and you get positional frequency response variation.

If you want to put it in terms of lobes, the whole room is in a single lobe below this critical frequency.

We never did get around to trying a delay and a rear set of speakers to counteract a wavefront (slapback was bad in an auditorium at college). My suspicion is that you could only do that at low frequencies such that a coherent wavefront hit the wall. In small rooms, I think the double bass array could work reasonably well, as long as the LPF was properly chosen. But maybe you could do it at higher frequencies if you had good pattern control from horn loaded speakers.



-tINY

[SAC]

Originally Posted by tINY

In the original drawing from Germany, the 4 woofers on the front wall are shown to be about 1/4 of the way in from the ceiling/floor and side walls. If you only consider frequencies below where the distance between the drivers is 1/2 wavelength, then most of the room sees no destructive interference from the multiple drivers (at least until the wavefront bounces off the back wall). Frequencies higher than that (where 1/2 wavelengths are less than the driver spacing) will not be coherent and you get positional frequency response variation.

I am very confused by this 'figuring'.

Any frequency where the spaced sources are greater than 1/4 wavelength results in comb filtering and polar lobing. I am not sure what 1/2 wavelengths have to do with this, except that they will result in comb filtering and polar lobing as well.

“1/4 distance to a boundary” and ½ wavelengths do not see interference until the signal ‘bounces off the back wall’? I'm not quite sure of the reasoning here regarding 1/4 distances and 1/2 wavelengths bouncing off walls..., but dependent upon the gain of the reflected signal (which will be less with significant spacing and the inverse square law) and the timing/distance, yes, the signals, regardless of frequency or total wavelength, will/can still act as virtual sources and combine via superposition and contribute still more to the total superposition.

Based upon the time or distance offsets, the initial null frequency and the null frequency interval at which the comb filtering will repeat is easily calculated.

With multiple spaced sources, you are subject to comb filtering and polar lobing due to the combination of sources from ALL spaced sources, be they real drivers, or virtual sources such as walls.
In this room we have the spacing from each driver to each boundary (up down, sideways) where the effective wavelength is 2X the separation distance – meaning that if the distance to the surface is 4 foot, then 2X = 8 foot. And 8 ft is the quarter wavelength corresponding to a 32 foot wavelength, or a frequency of ~35 Hz!

With a time differential of 7.1 ms.
And the initial null frequency is 71 Hz.
And the null frequency interval is 141 Hz, with the first three null frequencies at: 70.6 Hz, 212 Hz and 353 Hz.

If the horizontal and vertical driver spacing is, say, 6 foot between the 2 top horizontal speakers, then the 1/4 wave frequency is 24 foot, or ~46.9 Hz.
With a time differential of 5.3 ms.
And the initial null frequency is ~94 Hz
And the null frequency interval is 188 Hz; with the first three nulls frequencies at: 94.2 Hz, 282 Hz, 470 Hz.

Thus, the diagonal driver spacing would be 8.49 feet.
With a time differential of 7.5 ms.
And the initial null frequency is 66.5 Hz.
And the null frequency interval is 133 Hz; with the first three null frequencies at: 66.5 Hz, 199.5 Hz, 299.5 Hz.

The reason they are worse than with simply one driver or two closely packed drivers where the reproduced passband is within ¼ wavelength and all effectively sum is that you will have multiple ‘initial’ null frequencies! And you will have multiple repeating comb frequency intervals as well as corresponding polar lobing intervals. In other words, the direct signal propagated from the speakers will exhibit much greater comb filtering and greater polar lobing anomalies. And the more sources sharing the same offsets located on the same plane, the more the response is reinforced. This can become manifest in a significant impact upon the locational cues.


Combine this with the modal response, and you have a mess... MUCH 'larger' than if you did not have the multiple spaced drivers reproducing the same passband. Hence, far from being an optimal design, this configuration simply exacerbates the response anomalies in an attempt to increase gain. There are much better ways to increase gain without the attendant anomalies.

[tINY]

I think that you have a good start at the pieces. You are missing a few and not seeing the big picture.

You identified a couple of nulls below 100Hz, but they are only nulls if you are in the plane of the front wall (that your subwoofers are mounted on). If you do the calculations, you will find out that the nulls correspond to the 1/2 wavelength multiples that would otherwise form modal ringing - but you already conceded that point.

Then consider time of arrival from the 4 drivers to a receiver anywhere in the room where your head is likely to be (32-60" off the floor and 2 feet or more from a side wall). The path length differences will be a lot less than the ones you identified, which are along the 4 edges of the front wall.

I could walk through the math, but you seem to like to do that and my newborn needs a bit more formula and a fresh diaper. Then I need some sleep.

After that, think about the addition of the path length differences of the other 16 impulses from the 4 drivers' first reflections off of the walls and ceiling/floor if you are looking for how bad the comb filtering really is below 100Hz (or wherever you'd like to cross-over your subwoofer sub-system). I think it's not as bad as you imagine.



-tINY

[SAC]

I think you totally miss the point.

The point is that the arrangement is worse than with simply one driver or a better spaced pair. 4 spaced drivers on a vertical wall are not optimal, and the result is INCREASED comb filtering and polar lobing in addition to the modal response.

It is a comparative valuation. And one that increased over the presence of a single driver, whose behavior is confirmed in Pat Brown's comparative study.

And as far as path lengths, it is the separation between acoustic origins that matters as all signals originate on the same plane. The calculations are very easy and well vetted.

What is surprising is that you, for some reason, want to debate the existence of additional comb filtering and polar lobing that is a fundamental quality and direct consequence of spaced drivers, made worse by the additional drivers and boundary relations.

[SaSi_SiDi]

Spaced multiple drivers are not always producing comb filtering and directionality issues.

There is a configuration (Bessel array) that has been patented by Philips, where several identical drivers are lined up in an array (single or two dimensional) forming a source that resembles a point source.

It has been found that a 5 element array is the most efficient configuration in terms of cost and dispersion vs distance.

The above has nothing to do with the ongoing debate, just a point to show that multiple drivers are not by definition producing destructive interference.

In a small room we have a very sick situation. Sometimes the cure is another sickness, that when combined, removes all the symptoms.

I think we need to find the Harman Kardon white paper on subwoofer placement.

[tINY]

Originally Posted by SAC

I think you totally miss the point.

The point is that the arrangement is worse than with simply one driver or a better spaced pair. 4 spaced drivers on a vertical wall are not optimal, and the result is INCREASED comb filtering and polar lobing in addition to the modal response.

I disagree with your point. I don't miss it at all.

I am trying to explain to you why multiple woofers in small room can actually be much better. Part of that explanation is getting you to understand that the low-frequency behavior of a small room is nothing like free-field behavior. In small rooms, one woofer is among the worst options from a fidelity standpoint in the low-frequency range....

It is a comparative valuation. And one that increased over the presence of a single driver, whose behavior is confirmed in Pat Brown's comparative study.

And as far as path lengths, it is the separation between acoustic origins that matters as all signals originate on the same plane. The calculations are very easy and well vetted.

What are those calculations? And what do they say about positions within 15 degrees of perpendicular to the plane measured from the midpoint of the drivers?

What is surprising is that you, for some reason, want to debate the existence of additional comb filtering and polar lobing that is a fundamental quality and direct consequence of spaced drivers, made worse by the additional drivers and boundary relations.

Not at all. I simply question their relevance to the situation at hand. If those same 4 drivers were suspended in the middle of a hockey arena, then the effects you describe would be easy to measure and hear. When you turn a whole small room, effectively, into the throat of a bass horn, you see very different things.



-tINY

[tINY]

Originally Posted by SaSi_SiDi

I think we need to find the Harman Kardon white paper on subwoofer placement.

http://www.harman.com/EN-US/OurCompa...s/multsubs.pdf



Here is the link to Welti's very informative white paper.


-tINY

[SAC]

The Bessel array was mentioned twice earlier above. And it simply reinforces what I have said, as it is the ONE exception to the destructive comb and polar lobing interference caused by spaced sources! That means that the alternative spaced sources, such as what we have in the example in this thread, do NOT exhibit such a normalized response!

And the Harmon white paper is fine, except there are several very significant differences that are ignored.
You will note that it deals with woofers placed in the HORIZONTAL plane where ear height is assumed to never vary, NOT the vertical plane as is the system configuration cited at the beginning of the thread.

And as you will note in the white paper, there is a 'small' comment in frame 40 that despite everything they are doing - including their EXTENSIVE use of equalization and a mono bass management system to try to correct for the massive variations - there still results a soundfield where:

"the floor and ceiling modes would NOT be canceled, however these modes are much less important. Floor and ceiling modes do not cause variation in frequency response over the seating area, assuming ear height does not vary."

What is the floor and ceiling in their example is the front and rear wall in our example! And the horizontal plane in their example where ear height does not vary is the vertical plane in OUR example where ear height Does vary as we move closer or further away from the sources! Hence, while these problems may not be large in their example where ear height does not vary, as our example is rotated 90 degrees, where movement is not constrained, a horizontal shift does cause significant variation - and this IS a big issue as they note!!!! Thus the explicit limitation they note corresponds exactly to our situation!

And note the 'small' qualification noted in frame 39!!! "Room response between modes is IGNORED! How convenient, as this is precisely where the comb filtering and polar lobing initially become manifest in the ‘gray area’ of the cross over region about the critical frequency where both modal and specular reflections co-exist. The problem is that in our scenario that these variations ARE important, as we are effectively located where they say problems exist and remain! And the comb filtering and polar lobing are characteristics of the direct signal.
Their modeling is precisely oriented to focus on only ONE particular aspect of the response Modes.

And the model we are evaluating does NOT place the speakers in nulls and the listener in a null nor cancel out modes as their model seeks to do in the horizontal plane for optimal response! (panel 18)

Plus they are also applying LOTS of equalization to try to moderate exactly the variations that DO exist. And note their use of a mono bass management system that substantially simplifies the model behavior with the practical result of moderating significant variations from the model as presented.

Oh, and you keep saying that superposition and comb filtering and polar lobing applies only to a large acoustical space and not to small acoustical spaces by your continual reference to large sports and music halls. I really have no idea where or how you come to that conclusion. Apparently this is a result of your thinking that as the MEASUREMENT space was large relative to the bounded test area, that this does something aside from eliminating additional destructive issues causing more comb filtering and polar lobing.

But that is FUNDAMENTAL to measurements as they eliminate additional reflections in order to focus on the spaced reflections of ONLY the driver and the close boundary! To include the reflections of more spaced sources, the additional boundaries exacerbate more destructive interference!

Thus, removing the far surfaces simplify the model, resulting in a BETTER result, not a worse one!

Thus they have gone to great extremes to minimize the evaluation of their space for anything but modal behavior - as that was their stated constraint. And as such, there should be no surprise that their tests focus solely on the modal behavior of what is otherwise a very poor frequency response. (Despite their erroneous reference that a modal response is a reverberant field! ESPECIALLY in the LF response! But it does result in a persistence that is a resonance - something distinctly different from a statistically reverberant sound field! A modal behavior is the result of a STANDING pressure wave, which is the exact OPPOSITE extreme behavior from statistically random specular reflections! (frame 25)

But all in all, I suspect we have derived all the fun we can from this topic....

[SaSi_SiDi]

There is some (unqualified) testimony from individuals reporting that in (their untreated and unfavorable) listening rooms, mono sounds much better than stereo...

[SAC]

Originally Posted by SaSi_SiDi

There is some (unqualified) testimony from individuals reporting that in (their untreated and unfavorable) listening rooms, mono sounds much better than stereo...

In MANY cases a centrally mounted controlled dispersion(Q) mono source is OFTEN MUCH better in many respects!

[tINY]

Originally Posted by SAC

And the Harmon white paper is fine, except there are several very significant differences that are ignored.
You will note that it deals with woofers placed in the HORIZONTAL plane where ear height is assumed to never vary, NOT the vertical plane as is the system configuration cited at the beginning of the thread.

And as you will note in the white paper, there is a 'small' comment in frame 40 that despite everything they are doing - including their EXTENSIVE use of equalization and a mono bass management system to try to correct for the massive variations - there still results a soundfield where:

"the floor and ceiling modes would NOT be canceled, however these modes are much less important. Floor and ceiling modes do not cause variation in frequency response over the seating area, assuming ear height does not vary."

Please go back and read the premise of the study. Within the scope of the study, this is a resonable simplification (likely made to keep the presentation interesting to people other than those still reading this exchange).

You will also notice that they kept the woofers on the floor, placed woofers along the wall, and drove the identical woofers in-phase. This is how the majority of any commercially viable systems will be deployed.

Of course, you may not be interested in actually improving the system performance of sound systems sold to the masses, but that's what Harmon does. Specifically, they are looking to improve low frequency behavior in home theatres for multiple listening positions.

There is no reason these concepts can't be extended into non-rectangular rooms or into 3-dimmensional configurations where the woofers are deployed at heights other than on the floor.

The german concept, using front and back arrays driven with different signals, simplifies the problem by turning the room into a giant tube with cancelation at the back wall. This may be better in a lot of ways, but is hard to sell domestically (if you are married).

And note the 'small' qualification noted in frame 39!!! "Room response between modes is IGNORED! How convenient, as this is precisely where the comb filtering and polar lobing initially become manifest in the ‘gray area’ of the cross over region about the critical frequency where both modal and specular reflections co-exist. The problem is that in our scenario that these variations ARE important, as we are effectively located where they say problems exist and remain! And the comb filtering and polar lobing are characteristics of the direct signal.
Their modeling is precisely oriented to focus on only ONE particular aspect of the response Modes.

This is precisely why they included the study. Using a "RULE OF THUMB" is usually a massive simplification of a complex problem.

You'll notice in Slide 45 that they come to the same conclusion you do... Is there a point to bringing this up?

Plus they are also applying LOTS of equalization to try to moderate exactly the variations that DO exist. And note their use of a mono bass management system that substantially simplifies the model behavior with the practical result of moderating significant variations from the model as presented.

That is the point of using a mono subwoofer system with multiple, identical drivers. If you can reduce variation of frequency response over the seating area and then apply a couple of bands of equalization to normalize the response, you have improved the bass performance for the people trying to listen to the music or watch the movie.

I'm not sure what is so evil about moderating variations of the model - so long as the system you are designing can effectively employ the same techniques. Are you hoping for directional sub-woofer response in a small room?

Oh, and you keep saying that superposition and comb filtering and polar lobing applies only to a large acoustical space and not to small acoustical spaces by your continual reference to large sports and music halls. I really have no idea where or how you come to that conclusion. Apparently this is a result of your thinking that as the MEASUREMENT space was large relative to the bounded test area, that this does something aside from eliminating additional destructive issues causing more comb filtering and polar lobing.

But that is FUNDAMENTAL to measurements as they eliminate additional reflections in order to focus on the spaced reflections of ONLY the driver and the close boundary! To include the reflections of more spaced sources, the additional boundaries exacerbate more destructive interference!

Thus, removing the far surfaces simplify the model, resulting in a BETTER result, not a worse one!

I am not suggesting that the results of mulitiple time-of-arrival differences (lobing and comb-filtering) don't exist for low frequencies in a small room. I am suggesting that they are inconsequential when considered in such simple terms. All of the lobing and combfiltering you have alluded to have both constructive and destructive interference - this is the way all sound is in all acoustical spaces. It's the aggregation of all of the contributions that is important.

You still haven't applied your "better result" to model the relatively simple, practical application of the 4 woofers in the front wall of the german scheme. It's only 4 sources and 4 sources reflected off of 4 surfaces for a first approximation... 20 signals to superimpose. Did you do the calculations and find that it didn't support your argument?

But all in all, I suspect we have derived all the fun we can from this topic...

.

Until someone actually wants to put forth the effort to do some modeling or build 4 or 8 subs and take some measurements in a room....



-tINY