I need to cover my ceiling?

[kingrhythm]

Here's my room that I'm working on. I've added a lot of panels. All corners straddled w 6", ceiling corners straddled with 4" and walls covered with 2".
It seems that the more I add the more one resonant frequency rings out when I clap. Now with everything covered except for the ceiling, the frequency seems pretty loud and clean like a note.
I have thin carpet on the floor and nothing on the ceiling. I know that it's the bare ceiling causing this because the second I put something close to the ceiling then clap the ring disapears in that area.
So do I have to cover the majority of my ceiling and not just the section over my work area with the cloud? I just want to make sure this seems right or common?

Thanks

*edit*
Ok, I just read that this is called modal ringing so I have another question. My room is only 7 feet tall, and I have a cloud over my work area that is gonna be 8" down (4" thick and 4" gap) this is fine cause it's needed. But I need some place for tall people to be able to stand somewhat comfortably and I can't cover the whole room with 8" down. So given that my room is small and the thicker the better but I need to get rid of this ringing, which is better for the rest of the ceiling: 2" panel with 2" gap or 4" panel flush?

[Ethan Winer]

Modal ringing is different from flutter echo only in the range of frequencies affected. Corner bass traps will not help flutter echo that exists between two large parallel surfaces such as the floor and ceiling. You do not need to treat the entire ceiling, only the area where sound originates. (And also the loudspeaker reflection points which is a different issue.) So instead of clapping your hands, play a hand clap through your speakers. If you don't hear flutter echo you don't need additional absorption. However, a lot of people record things in their control room, so you probably need to treat the ceiling over that area.

--Ethan

________________
The Acoustic Treatment Experts

[kingrhythm]

Thank you for the explanation Ethan! I do record a few things in here so I'm gonna cover it. It's really interesting how acute it got, even when I talked just a little bit above normal talking level I would hear it subtly ring out.

[Glenn Kuras]

Originally Posted by kingrhythm

Thank you for the explanation Ethan! I do record a few things in here so I'm gonna cover it. It's really interesting how acute it got, even when I talked just a little bit above normal talking level I would hear it subtly ring out.

I see rooms all the time without treatment on the ceiling and it always makes me wonder why. It really is a critical area that is over looked.

[AnthonyRochester]

Originally Posted by Glenn Kuras

I see rooms all the time without treatment on the ceiling and it always makes me wonder why. It really is a critical area that is over looked.

In my room I have very effective broadband absorption on the side walls, and it isn't physically practical to treat the ceiling because I am in a rental place. But it seems the side wall absorption seems to be enough to bring the first reflections down to a low enough level..

[SAC]

There seems to be a misunderstanding of the psycho-acoustical issues underlying the purpose of treatment.

Additionally, this treatment and the requirements at hand are dependent upon the room's intended use.
If the room is a non-critical space - meaning that it is a live or recording room where one wishes to tailor a particular sound, the comments below do not apply, and the comments above referring to incremental improvements are indeed valid.

But, if the space is intended for critical listening, as in a control/mix environment or a critical listening room, the below issues are critical to accomplishing the intended benefits that are commonly associated with an ISD/RFZ gap. And the cited reference levels are conditions included in both the physics and psycho-acoustics for which purpose the gap is designed. Thus we need to be both aware, and have a means by which to insure such conditions are met in order to realize the full capabilities of the design feature.

What follows makes no attempt to address later arriving energy which is beyond the intended scope of this post...

The first is the sufficiency regarding the treatment of early arriving reflections.

It is Not a matter of simply trying to damp 'most' of the reflections in a defined ISD gap. All energy arrival within the ISD/RFZ gap, be they reflective or diffractive, must be damped a minimum of 20 dB SPL below the direct signal level. (ideally 30 dB SPL! No small feat!)!

(And please don't ask me what this is in 'db FS'! Who knows? That is a limitation of the measuring platform used - one that does not exist in the larger more comprehensive platforms. For our purposes dB SPL is the useful unit)

Failure to adequately (in terms of gain) damp ALL of the reflections in the ISD region still results in the loss of intelligibility and errors in imaging and localization. This is not a matter of getting it 'mostly' right. And reflections arriving in this period off the work surface and/or ceiling render all of the treatment that may exist on the side walls moot.

Additionally, while one can make gross generalizations as to what treatment panel size is 'adequate'. Guesses, feelings and generalizations, while convenient, mean little. What I mean by this is on occasion a smaller panel may be adequate. More generally, the larger panel will more likely to be adequate. But the size is not what ultimately makes one more adequate than the other! Matching the treatment to the actual behavior is what is critical.

And this is why the convenient methods such as mirrors, etc. fail. Oh, they give a good generalsome of the reflection paths. But as is mentioned above, 'some' or 'most' is not adequate to address what happens psycho-acoustically in an ISD. That is why measurements are important.

The ETC can tell you if ALL of the indirect sources are damped, and if they are damped adequately. They can tell you if you are still getting sufficient reflections off the absorber surfaces to render them less than adequate. You can still intercept all of the reflection sources and still have absorptive panel reflections that exceed the minimum required levels.

Thus, treating the side walls and neglecting to appropriately and effectively treat reflections off the work surface and ceiling (and any other source that might exist in a particular room) accounting for ALL of the signal sources arriving withing the ISD/RFZ interval fails.


There is much more to the proper determination and establishment of an ISD gap and its termination (yes, the termination is fundamental to the ISD/RFZ gap with laterally arriving energy returns NO MORE than 10-12 dB SPL lower than the direct signal) and what one may do with a later arriving sound field, but that can be discussed elsewhere.

[AnthonyRochester]

Originally Posted by SAC

Failure to adequately (in terms of gain) damp ALL of the reflections in the ISD region still results in the loss of intelligibility and errors in imaging and localization. This is not a matter of getting it 'mostly' right. And reflections arriving in this period off the work surface and/or ceiling render all of the treatment that may exist on the side walls moot.

from my experience, I'm not sure that the above is a fair statement.. Just treating side walls gives a really big improvement, and its not like I'm unable to do any decent audio work if I don't treat the ceiling too..

[SAC]

The Henry Precedence Effect and the Haas corollaries have objective determinate criteria defined by time and gain relationships which are not simply guidelines. And while their satisfaction most definitely contributes to subjective perception, they are dependent upon the satisfaction of the objective criteria.

You may subjectively prefer an environment, but the accuracy of the imaging, localization and intelligibility is not subject to one's subjective preferences any more than, for example, equalization may make things 'sound better', or one may find watching television without their glasses adequate based upon their subjective preferences - but one should not confuse purely subjective preference with what actually resolves fundamental limitations regarding the accuracy of perceptual cues. To use a really poor analogy, it may feel plenty cold at 40 degrees, but if one is expecting water to freeze, the temperature is going to (other variables precluded) need to reach 32 degrees. 40 degrees simply won't do it.

If one simply wants to make something 'feel better', one can do whatever they like. But the goal is not simply to satisfy subjective preferences. If one seeks to objectively improve psycho-acoustic qualities such as the accuracy of imaging and localization, the intelligibility of the arriving signal, as well as translation accuracy, one satisfies the requirements for such behavior in terms of the time and gain relationships of the arriving signals, or they do not. Such time and gain relationship criterion are objective, not subjective. And the actual benefits in terms of accuracy, while it may lend itself to improvements in the subjective realm, are objective in nature.

What folks need to realize in these models is that miscellaneous reflective and diffractive surfaces such as work surfaces, speaker mounts, ceilings, etc., if left unresolved and are not treated effectively to the same degree as other reflective and diffractive sources, are the equivalent to early energy control as flanking paths are to transmission/isolation control.

The effectiveness of such treatment is determined by the weakest link, NOT by the most effective treatment!

[AnthonyRochester]

But successful improvement in the accuracy of a listening room isn't always either complete or incomplete. It seems too much like a black or white view. You can install some treatments that improve the room, and then do more things that improve it a bit more. Am I wrong?

[SAC]

Yes and no.

Remember, what you say is quite accurate for a live room where one is creating a particular response.

But in a critical listening environment, such as a control/mix room, we are using the ISD to effect a particular psycho-acoustic response that specifically enhances imaging, localization and intelligibility. Here the requirements are much more strict, as we must meet certain objectively defined criteria in order to effect the psycho-acoustical response described within the Henry Precedence Effect and the Haas corollary.

In some cases incremental improvement results in real benefits. A good example is the live room, where we are in effect creating effects. In other cases, limited by the 'weakest variable', incremental improvements do not achieve the desired result until all conditions are satisfied. Such is the case with the ISD in a critical listening environment such as the mix or control room.

And here we must meet a set of minimum objectively defined and verifiable criteria.

Sure, aspects such as bass traps can effectively address modal behavior, but they do little to nothing to remediate the substantial issues arsing from specular issues. And a later arriving diffuse field can be 'added' at a later time in order to augment the total response without detriment to the fundamental ISD.

But adding treatments without completing the necessary elements of a particular 'performance module' (pardon the use of some rather squirrely terms) such as the ISD, does not result in a significant improvement - as the conditions necessary to modify the Henry Precedence Effect and Haas corollary are not satisfied. thus an ISD/RFZ gap that is not sufficiently constructed where ALL of the energy within the region is not reduced in gain will NOT result in an effective ISD gap. This is a case where incremental improvements are not sufficient to modify the perceptual issues defined in the Henry Precedence effect and Haas corollary.

One might think it akin to building a house. Erecting a foundation and the walls is definitely a step toward the fulfillment of the goal and provides a really great subjective sense of accomplishment; but without a roof, one can hardly say that said steps are effective in achieving a weather-tight structure - a fact that may be reinforced in the rain... Thus in the rain you are no drier then before all of the hard work as all of the necessary conditions have not been satisfied.

Or to pull another lame analogy out of hat...heating water may make it subjectively VERY hot, but unless you heat it to 100 degrees C (STP), it 'ain't' going to boil! And all the subjective satisfaction in the world won't change the fact. And in the acoustical sense, in order to effectively invoke the principles, it has to boil, not simply be hot.

An airplane may try 'really hard' to fly, but unless it achieves a certain 'objective' velocity (assuming lots of other satisfied objective variables!) it's not going to fly - despite how subjectively satisfied one might be with the attempt or how many additional variables have been met!

Some things in life accommodate progressive improvement, and any additional effort results in objective incremental advantage. Some things, like in quantum mechanics 'don't care' how much work is expended in the process of attempting to satisfy a goal.

(In fact, in quantum, if one is not aware of this fact, energy is 'quantized'. For instance, assuming that macro water acted on a micro scale, to boil water requires the application of precisely 100 degrees C - to apply 99.9999999999999999C or a billion or 100 billion C will not boil the water. It must be precisely 100C .) The effort doesn't matter - the satisfactory conditions must be met before a result is achieved.

It's a physics and psycho-acoustical thing - while one's perception of the event may be subjective, the criterion to satisfy the condition to which we refer here is objective. There are critical points that must be satisfied - and it doesn't care to what degree someone is in touch with their feelings.

Like sound isolation, building the world's 'best' most isolated wall may be a step, but without addressing the flanking paths and raising their performance to an equivalent standard, the wall, however subjectively and objectively amazing and rewarding as it may be, is alone ineffective in solving the transmission issue.

And until ALL necessary objective elements are achieved, the results will still be limited by the weakest link. And while the efforts may be subjectively satisfying, objectively - in terms of the goal - the performance is still equivalent to where you were at the beginning.

And in this case, without fully addressing all energy arrival in the ISD/RFZ gap, and without satisfactory termination of the same, you will NOT trigger the Henry Precedence Effect and Haas corollaries in the manner many incorrectly assume are satisfied. You will still have the imaging, localization and intelligibility issues that existed prior to the intermediate treatments!

So my point is not to give up or to not try! - the message is to not think that your subjective satisfaction satisfies the real objective goal that drove you to begin the process! And that process requires that certain objectively verifiable conditions be met. Thus it only makes sense to understand, acknowledge and then meet these objective criterion if one's goal is to satisfy and invoke the effect.

And the ISD is one area where extraneous energy arrival renders all of the incremental work for naught. Thus folks need to be aware of this, as so many are trying so hard to achieve a level of performance that all of their effort deserves! It is a shame for some to work so hard and then to stop just short of achieving their goal but for a lack of awareness of the necessary criterion!



Geesh , how many goofy analogies and similies can be fit into one post?...

[AnthonyRochester]

OK OK I get your point....

[SAC]

The point is, the Henry Precedence Effect and Haas corollaries have critical thresholds that must be satisfied in order for the perceptual gains to be realized.

And the introduction of any signals exceeding the gain thresholds during the interval changes the effect and invalidates the desired behavior. In other words, what they think they have achieved has in fact not been achieved.

The problem that is more often then not happening is the common, but erroneous, assumption that simply 'getting close' is sufficient to satisfy the requirements of the principles sufficient to elicit the response. It is not.

Like a photography dark room, it is not sufficient to eliminate most of the 'white' light. Just one quick flick of the light switch or an open door squirrels the entire operation - regardless of how 'dark' the room is for the remainder of the time. The same with the ISD gap. And in this case, ALL energy during the gap must be below 20 dB (...preferably 30 dB!) to successfully elicit the perceptual effect. And this effect is optimized by the termination of the gap with a laterally arriving signal (preferably diffuse) arriving at not more than 10-12 dB down in gain from the direct signal. Less, and it too fails to be effective in eliciting the desired effect.

Thus, there are good reasons why such extreme steps as the elaborate geometric designs of P. D'Antonio's RFZ geometry have been employed to establish an effective ISD/RFZ gap! It is not simply because folks had extra time and materials and simply thought it would look cool. While it is but one method, the mechanics of a truly effective ISD gap are not trivial.

And the ETC response allows one to examine, identify any anomaly that may compromise this design feature, as well as to verify the fulfillment of the design requirements necessary to satisfy the function of this structure.

The ISD gap is effectively anechoic. And this definition does not mean 'kind of' or 'almost' anechoic. Thus ALL of the energy paths occurring in this interval, be they diffractive or reflective, MUST be addressed adequately. That is simply what is required to make it work. But quite frankly, and unfortunately, what many casual users are satisfied with is NOT a true ISD gap. And the unfortunate part is not thay the room may not qualify for a plaque or some symbolic indication of its satisfying the criterion, but rather that the actual response of the design is not as effective as it should be! Instead of a real substantial 'performance boost', one has simply invested time and money for a few fancy flame decals expected to make your car go faster. And THAT is the real problem.

Note, all we are addressing now is the gain level required during the ISD gap. We have not addressed the optimal length (and no, there is not simply one 20 ms optimal length as is so often cited!), nor how to determine a usable time period of the ISD gap in the space. So there is still more to this issue.


I would much rather folks understand this and treat the space sufficiently and completely, instead of doing allot of work satisfying 'most' of the necessary criteria ultimately for very little real gain. A little extra attention to detail with respect to the requirements will result in real improvements beyond what one is obtaining currently.

[cborg]

Originally Posted by SAC

The same with the ISD gap. And in this case, ALL energy during the gap must be below 20 dB (...preferably 30 dB!) to successfully elicit the perceptual effect. And this effect is optimized by the termination of the gap with a laterally arriving signal (preferably diffuse) arriving at not more than 10-12 dB down in gain from the direct signal. Less, and it too fails to be effective in eliciting the desired effect.

It may all be very good but how do you accomplish that in a small room?
(Small room meaning the longest dimension being maximum 4-4,5m)

[SAC]

Here's a deal.

I started jotting some points and principles and techniques down, and then realized that, after taking up most of the morning doing something totally foreign to what I am SUPPOSED to be doing I also realized that a slightly more comprehensive approach to the subject is required.

Rather than just give a few suggested tricks and methods - leaving some to question why this is even important, or trying to explain the theory behind the effect that can be achieved, leaving others to wonder 'so how can you achieve it (such as your question), and trying to answer the questions without delving into a few more of the real and practical issues often encountered as well as how each can be evaluated and resolved...

...Confused yet? I found myself with a page or two with all sorts of tangential loose ends referring to all sorts of important aspects trailing all over the place that required a bit more fleshing out in order that someone might actually read it and come away with an understanding of the 'theoretical' underpinnings - the 'Why' we do it- while also achieving an appreciation for a few practical avenues of 'How it can be done' as well as 'How can one identify, analyze and verify the elements as well as the effectiveness and success' of each step in the endeavor.

So - rather than post something that in answering a few issues out of context leaves just as many questions unanswered, allow me to spend a bit of time coming up with a reply. Already I see this becoming a text in acoustics - which I have no time nor desire to do, as each issue segues into related issues - so some restraint in scope is going to be necessary! But let me try to come up with a bit more on the ISD itself, and how one can make it work in a small room while addressing the other issues that one commonly encounters.

But let me say at this point, that achieving a desired end is MUCH easier when one begins with a clear image of the desired goal, and THEN chooses treatments that will, at every step and purpose, contribute to the end result, rather than simply starting with a bunch of general treatments, applying them, and then wondering why one cannot achieve a goal is simply not supported supported by a 'death of many cuts' due to many individual choices made without regards for the ultimate desired response.

And then we get to listen to many tell us that the goal is simply not possible.

In some cases partial compromise will be made, especially considering the many closet type rooms some are limited to. but a reasonable sized small room can be made to function properly, and a 4 x 5 meter room is not 'overly' unreasonable.

But give me a week or two - or 6 months to a year (sorry, I'm -hopefully - kidding) as things are becoming pretty crazy both with work and with the so called home life - especially with the advent of some welcome relief to the oppressive heat.

And with a bit of luck we can explore the ISD as well as some methods that can be utilized to effectively implement it to its full potential - or at least to the greatest degree possible - in a small room.

[cborg]

Well, I didn't mean to force you to go write an RFZ/ISD acoustics book.

I am interested in any solutions/examples regarding the RFZ/ISD mentioned above for small rooms though.

From what i can gather regarding RFZ in small rooms is that you either have to treat a big percentage of the surface for lowering the relative level of the early reflections as well as decay time and modal response. And even the back wall may reflect too early making it difficult to preserve enough energy for later (diffuse) arrival.
Or the zone will be very and perhaps too small.

Thats why I'm wondering.

[SAC]

No. If you follow the notion of using a mirror to identify simple ray tracing vectors, you will end up treating just about the entire room! And all that accomplishes is to create a dead room - after which folks complain that there is insufficient energy to adequately terminate their anything but anechoic ISD gap - all the while blaming the size of the room and not their choice of methodology of indiscriminately using too much 'non-specific' absorption!!

Unfortunately, imprecise casual tools such as mirrors tell you nothing about the arrival time nor the gain. And they are incapable of identifying the very near to speaker diffractive and reflective sources.

And in treating the room in the manner you state, you are assuming it all has to be treated. That is simply not the case.

In using the ETC, measure the actual energy dispersion that addresses many more real world factors! Speaker polar dispersion, boundary impedance, etc., are all incorporated into the real world picture the ETC provides. You are able to identify every energy source, their relative gain and precise arrival times (assuming your platform displays the dB SPL scale! and that you employ loopback propagation delay correction in the configuration), and you resolve EVERY energy source to its actual incident region, be it on a room boundary or a speaker ledge or cabinet edge. The ETC allows you to dissect the grunge and to surgically remediate them without overdamping the space.

You then surgically treat the reflections. And you use the ETC to verify the effectiveness without overdoing the absorption!

Also, you use frequency selective bass traps, which, if they are porous corner traps, you employ truly reflective front coverings - not merely film that may only reflect low energy content high frequency energy. And you orient the front facing to redirect this energy outside the listening position to the back of the room - thus preserving as much finite energy as possible.

In other words, you surgically treat what is necessary to create the early arriving anechoic (down >= 20 dB SPL) ISD/RFZ region. No more. And you ferret out each specific source and treat it - you do not simply treat the entire room because it is possible the energy Might be reflecting there! The ETC provides you with VERY precise verifiable path information!

So, your first concern is creating a truly damped ISD gap. Not almost, not kind of. As it is my assumption that you want to satisfy the requirements necessary to validate the Henry Precedence Effect and the Haas corollary. And any spurious energy that exceeds the 20 dB SPL gain level relative to the direct energy effectively terminates the gap.

But then, if it is at all possible, we also want to properly terminate the ISD gap - not with some intermediate insufficiently damped low level spurious energy source that simply screws up the imaging and localization and intelligibility as does a spurious signal referred to above. Instead we want to terminate it in such a manner that it reinforces the Henry Precedence Effect and Haas corollary and effectively lock in the image and localization. And remember, while one can state some very general approaches, the actual treatment is dependent upon the realities and specifics of each actual room.

And in order to achieve this augmentation, the significant return must be no more than 10-12 dB SPL down from the direct signal gain.

Ideally, in order for the later arriving soundfield to most seamlessly blend with the direct soundfield, we optimally prefer a laterally arriving diffuse soundfield.

And if the room is large enough (or if we are lucky enough to be able to employ techniques such as a coupled space to generate a diffuse field with devices like space couplers), then great, we can proceed with developing that process.

But I am assuming that the room is not large enough to achieve a sufficiently diffuse energy field.

But we can STILL terminate the ISD and gain the psycho-acoustical benefits with regards to imaging and localization that the termination of the ISD provides. And we can do it with specular reflections.

One can argue that it is not as 'good' as with a diffuse energy return - but we are not blending a later arriving diffuse energy field. And it is distinctly superior to an unterminated ISD which lacks the reinforcement of the imaging and localization that comes specifically from the termination of the ISD.

How do we do this? Very simply, we redirect the energy that is not incident upon the listening position. This includes ALL energy that is not specifically required to be damped - meaning you don't indiscriminately apply absorption everywhere! And you need not apply absorption to the back wall as long as the energy is adequately managed and redirected appropriately. So, rather than the 'usual' absorption or the common diffusion, we use reflection as a treatment on the back and rear side walls and corners to direct and manage a large energy return. (and if we have sufficient energy of sufficient gain to return, laterally mounted 1D QRDs/PRDs can be used to generate a diffuse return - otherwise we will return relatively high gain specular energy.)

The result is a fully developed and functional ISD with termination sans a later arriving diffuse field. We gain the significant advantage of a fully functional ISD that has distinct advantages over a 'dead' space, while we still lack the 'advanced' advantages of a fully developed later arriving diffuse soundfield. It is the middle ground that is most acceptable in a small space that will not support a later arriving soundfield.

All energy not impinging on the ISD is redirected to the back of the room where it can then be reflected to the rear lateral locations where it is reflected back to the listening position by what have been referred to as "Haas kickers".

For instance, depending upon the measurements, gain and exact topology of the room, energy can be directed, say to the back wall where by the use of angled reflectors we can redirect said energy to the rear side walls, where either by use of 1D diffusors we redirect the focused energy to the diffusors and the high gain energy is redistributed to the listening position.

But if the signal gain is not sufficient to meet the gain requirements coming from a diffusor (as, while made of reflective materials, they exhibit much higher than hoped for signal losses), we simply employ reflective panels to redirect energy of sufficient gain in a focused manner back to the listening position.

And a room of say, 5 meters deep, as was mentioned in the example, with the listening position of say 6 feet from the front wall, will support a direct straight-line ISD, front front straight back to the back wall of ~ 24 feet, or 21.2 ms. But if we are creative, we can redirect that energy to the rear side walls, or even the corner (with bass traps installed!) for reflection and increase the travel distance while orienting the energy return to the rear lateral positions and increase the travel time sufficient to terminate the ISD with a sufficiently strong energy return.

And this is all rather simply able to be accomplished with a series of reflectors on the rear and side walls and/or corners. They can be adjusted and their orientation and effectiveness verified with the ETC. I have even seen concave reflectors that focus and redirect the reflection incident on the back walls over to the rear side walls (or corner) where either a Haas kicker/reflector, or if the gain is sufficient, a 1D QRD/PRD can be used to redirect either the focused or diffuse energy back into the listening area and effectively terminate the ISD. Your options are conditioned by the actual topology and circumstances, but there are more than a few ways to effectively affect the acoustical response sufficient to objectively meet the requirements and to subjectively trigger the full benefits of the Henry Precedence response and the Haas effect resulting in vastly improved imaging, localization and intelligibility.

You may not have the benefit of an exponentially decaying laterally arriving diffuse energy field adding a sense of space and envelopment, but you WILL experience greatly enhanced accurate imaging and localization that is afforded by an optimally tuned ISD.

Note: This technique is specifically referenced in both Sound System Engineering and in The Master handbook of Acoustics.

And, if circumstances change, or one later wants to develop the later arriving soundfield (assuming the potential and the opportunity exists), a diffuse soundfield can always be added. Thus, even in a larger room, the prime concern is to fully develop the ISD with proper termination. Then, as resources or the environment allows, the room response can be further enhanced by the addition of a diffuse field.

There is much more to this as many techniques have been quickly been skimmed over - and methods for determining the optimal ISD of a control/mix environment relative to the response of the live room, but that is a bit beyond the scope of this post - which is already too long! ...sorry...

[cborg]

Thanks.
As I see it there are still compromises when the room is small but I wont hijack this thread anymore either but instead we might open another one for this.
It'll be interesting to read and discuss what could be done and what solutions are available for the serious home builder.
PS. When I say treatment i don't imply absorptive treatment exclusively.

[DrFrankencopter]

Originally Posted by SAC

Unfortunately, imprecise casual tools such as mirrors tell you nothing about the arrival time nor the gain. And they are incapable of identifying the very near to speaker diffractive and reflective sources.

You can conceivably estimate arrival time with a piece of string taped to the the mirror and your speaker. Really what you are about is what is the incremental time over the direct sound. If that's over 20ms or so, then don't kill that reflection. A crude way to do this would be to say 20 feet = 20ms, and add the length of the direct path to it. If a reflection path back to the listening position is shorter than this, then it's a candidate for absorption.

If you think about what this implies physically, it shows the importance of considering the work surface (console splash), ceiling, and floor paths. It shows pretty much immediately that monitor bridges are recipe's for trouble. ETC will give you a yardstick for how well you're doing at addressing the early reflection paths. Remember too that deflection can be an option.

Regarding ISD for small rooms....well, how small is small? I'd think that if the listening position is less than 10 feet from the rear wall, it's gonna be trouble, and would imply that a dead rear wall might be in order.

Cheers

Kris

[SAC]

The only problem with generalizations is that they are generalizations.

There is Nothing magical about 20 ms. That figure comes from the study of concert halls in large acoustical spaces.

It varies with each small space. Each room has a characteristic delay gap. It is the process of design and treatment which simply renders this gap effectively anechoic (or "reflection free")! And a discussion of this is another topic.

And gain and distance?

One issue , a mistake rather than error, that is only an issue in the manner in which some results are displayed, is that the time from direct signal arrival to the arrival of successive energy 'reflections', tells us far too little in terms of attempting to identify the path. It is akin to trying to identify an unknown path (say, from LA to NYC) by only having the distance between Poughkeepsie and NYC.). I might suggest Phoenix or Dallas or Salt lake City or Spokane might be acceptable options. And heaven knows, they're all the same, right? ;-) The fact is, its insufficient information. But again, to the degree that it is not a necessary error - in fact it is not an error at all but a mistake - and that falls under either a limitation in the platform or in operator error.

From the generalizations, we should see near continuous distribution of specular reflections with near equal energies for each boundary plane where there is incidence. Thus there should be a tightly grouped cluster of reflections gradually diverging relative to each identified point of incidence until the angle of incidence reaches a point where the energy is no longer convergent upon the listening point. This also assumes a near omni dispersal characteristic for each speaker source. (And they are? Aren't they? ;-))

Reality says that we come no where near to recording such a behavior for a variety of reasons.
So, which point of incident is the dominant 'source' with maximum gain? And what is the gain?
And what is the spatial character of the 'reflected' energy that directly impacts the gain and the time distribution?

The problem is that the generalized ray tracing model fails to accurately predict the real world behavior. Its a great conceptual demonstration tool, but it fails to provide sufficient detail to analyze behavior for which our senses are unable to resolve, desopite the fact that said behavior has a very real effect upon our psycho-acoustical experience.

And if we only have the difference in arrival times between the direct signal and the reflected signal, the full path cannot be determined. Thus the total time of travel from source to mic for every path must be obtained, not simply (as we keep seeing) from the direct arrival translated to equal 0ms in the time scale.

The string method is an 'acceptable' method to 'reverse engineer the path once a specular energy wavelet is identified. But it fails to reliably predict and discriminate between the possible and actual occurrence of the energy arrivals. And therein lies a fundamental limitation if one is attempting to surgically treat Only the real problem while simultaneously trying to maximize and manage and redirect the finite amount of energy that remains.

If it were not so, we would not see the predominate response in small spaces be the dead room. Instead the feat of cultivating and managing a substantive later arriving energy field would be the norm - even trivial - and we would instead simply be instructing folks as how one can redirect it productively.

The bottomline is that it was the precision measurements, in particular the ETC, that made analyzing this process simple. Literally.

Thus the devil is in the details - referring to the creative deployment of strategically based treatment strategically damping only that which requires damping (and note the development of the RFZ configuration which uses reflection to avoid the additional absorption of the finite energy available while redirecting the energy around the listeners ISD in order to more effectively preserve the energy for use in a later arriving exponentially decaying laterally arriving diffuse energy field!).

To identify not only a possible path, but to be able to determine the relative time and gain and spatial and temporal distribution of the energy is necessary to strategically treat it - if treatment is even necessary! And for this the energy-time measurements are indispensable. As the goal is to minimize the amount of absorption necessary in a small room - NOT to provide excuses to use more!

And there is good reason to acknowledge what Everest documents in the Handbook of Acoustics, that the use of the ETC measurement is critical to the development, analysis and subsequent understanding of this behavior in his chapter covering the Acoustics of the Control Room. And even he states that "The energy-time display ... gives hope that small listening rooms of genuine quality are now a possibility." (p.300 Hndbk..., 2nd Ed)

Oh, and to cite another example that pushes the generally imagined constraints... Helmuth Kolbe has built several VERY shallow rooms featuring, not only a defined and terminated ISD, but a full LEDE response with a diffuse sound field in a room not much deeper than 10 feet. (E.g.: the A+D Studios in Switzerland pictures in Everest, figure 17-16 of the 2nd ed.) And creative design has enabled full blown LEDE response rooms to be installed in mobile trailers which barely enabled a full size console to be installed. (PM me if you need a picture and description) So, the room itself is not necessarily the limiting factor - the role of the designer and their creative use of objective technology both for analysis and strategic treatment in order to meet objectively definable goals is a factor as well.

[DrFrankencopter]

Originally Posted by SAC

There is Nothing magical about 20 ms. That figure comes from the study of concert halls in large acoustical spaces.

Isn't 20ms approximately the end of the Haas zone where echoes start to be perceived as discrete? I'd think this would be a number to shoot for (or greater).

Cheers

Kris

[SAC]

In a large acoustical space, yes, 20 ms is the generally accepted target ITD figure. But you can trigger the response in periods as low as 8-9 ms via the proper design of an initial signal delay (ISD) gap.

The additional aspects of the Henry Precedence Effect, of which the Haas effect is but a small corollary, can be manipulated to trigger a variety of imaging and localization effects. And the use of such criterion in conjunction with an effectively anechoic (reflection free/20+ dB SPL) ISD region based upon the individual characteristics of the small acoustical space (admittedly, a 'longer' period of time is 'better' - up to a point) combined with the high level terminating return (max 10-12 dB SPL down from the direct signal), can effectively 'trigger'/induce the desired psycho-acoustical response.

Note, the ISD is the small acoustical space 'variation' on the large acoustical space concept based on the research of Leo Beranek and the term he coined for the phenomena which he called the Initial Time Delay (ITD) gap.

The origins of much of what was determined with regards to the ISD was a result of the psycho-acoustic research of Puddie Rodgers, who noted the similarities between the comb filtering patterns created by misaligned sound sources and the comb filtering created by the folds of the outer ear (pinna) and reflections off the torso. (Pinna Transforms & Sound Reproduction, AES Jrnl vol 29 April 1981)