Speaker positioning and setup.

[bdiament]

Hi Jud,

 

I am thinking there has got to be academic research on such topics as human auditory system reaction to direct/first arrival vs. reflections.

 

More than half a century ago, Helmut Hass talked about localizing sounds based on arrival time.

 

Best regards,

Barry

Soundkeeper Recordings

Barry Diament Audio

[Fokus]

I am thinking there has got to be academic research on such topics as human auditory system reaction to direct/first arrival vs. reflections.

 

The first 30% of Toole's book is a summary of exactly that.

[Bob Stern]

I am sorry, but at a fundamental level that is simply not true. There are no separate time and frequency or amplitude domains. They are just different ways for looking at the same underlying reality.

 

A room mode is a minimum phase phenomenon, which is math-talk for saying that its temporal/phase and frequency behaviour are tightly coupled. Change one, you change the other, and in a predictable way. As such a room mode's boost in the frequency domain and its temporal overhang can be perfectly corrected by a matching minimum phase equaliser.

 

 

Your statement is true only for the lowest frequency room mode, i.e., the frequency at which the longest dimension of the room is a half wavelength. For frequencies at which the propagation distance, including wall reflections, from the loudspeaker drivers to your ears is greater than one wavelength, the behavior is not minimum phase.

[Bob Stern]

Side Wall Reflections

 

My speakers are at the 1/5 positions along the long wall. I toe in the speakers to point directly at the listening position. The first reflection point on the side wall then is behind the plane of the front baffle of the speaker, i.e., more than 90 degrees off the speaker axis. I assume this toe-in should prevent the side wall first reflection from being a significant problem.

 

Barry, does this seem reasonable?

 

(I imagine the first reflection off the ceiling is my biggest problem. When my wife has a weak moment, I want to try putting a foam square at the ceiling first reflection point.)

[mayhem13]

Side Wall Reflections

 

My speakers are at the 1/5 positions along the long wall. I toe in the speakers to point directly at the listening position. The first reflection point on the side wall then is behind the plane of the front baffle of the speaker, i.e., more than 90 degrees off the speaker axis. I assume this toe-in should prevent the side wall first reflection from being a significant problem.

 

Barry, does this seem reasonable?

 

(I imagine the first reflection off the ceiling is my biggest problem. When my wife has a weak moment, I want to try putting a foam square at the ceiling first reflection point.)

 

The first reflection points would be in relation to the speakers directivity.....as to the ceiling, many speakers exhibit poor vertical directivity in the upper mid/treble area....but LF content .....well that's another story. You won't be absorbing or diffusing any of that without extraordinary methods. There's some excellent designs that deal with the ceiling problem by using a woofer above the mid and high drivers...represented by W and the mids by M and tweeters by T....in a vertical alignment represented as WMTMW. Other that this, you're stuck with floor bounce and an extremely out of time ceiling bounce.

[Jud]

...as to the ceiling, many speakers exhibit poor vertical directivity in the upper mid/treble area...you're stuck with...an extremely out of time ceiling bounce.

 

A/k/a "soundstage."

[bdiament]

Hi Bob,

 

Side Wall Reflections

 

My speakers are at the 1/5 positions along the long wall. I toe in the speakers to point directly at the listening position. The first reflection point on the side wall then is behind the plane of the front baffle of the speaker, i.e., more than 90 degrees off the speaker axis. I assume this toe-in should prevent the side wall first reflection from being a significant problem.

 

Barry, does this seem reasonable?

 

(I imagine the first reflection off the ceiling is my biggest problem. When my wife has a weak moment, I want to try putting a foam square at the ceiling first reflection point.)

 

Unless you sit *very* close to your speakers, I would not agree that the first reflection on the side wall is behind the plane of the front baffle of the speaker. (Even if you sat a foot away, I'd still think the reflection would occur somewhere between the plane of the speakers and your listening position. I think you'd have to be *behind* the speakers for the early reflection points from the side walls to be behind the speakers.) In my experience, side wall reflections are still an issue, regardless of toe-in. That is to say, when the reflection points are properly treated, the benefits are clear (pun intended ;-})

 

An easy way to find out the locations of the reflection points is to enlist the aid of an assistant to hold a small mirror against the side wall, at your seated eye level, with you in the listening position. Have the assistant walk along the wall, holding the mirror against the wall until you see a reflection of one of the speakers. Mark the extent of the reflection. Have the assistant continue until you see the reflection of the other speaker. Mark the extent of that reflection too. (Note there will be one reflection per speaker, per room boundary and that the reflection points will not move as speaker dispersion changes, as they are based on *your* position and the *boundary* position relative to the speaker position.) Those are the areas that need absorption.

 

I'm ready to bet that neither of these is going to be anywhere *except* between the plane marking the front of the speakers and the listening position.

 

The same will occur on the ceiling. Ideally, I'd want to treat the ceiling reflections too (remember, there will be one per speaker per room boundary). However, my experience has been that unless the ceiling is unusually low (or the speakers unusually high), this won't have anywhere near the effect that wall reflections do. (A carpet or rug takes care of the floor reflections.)

 

As always, please do *not* simply take my word for any of this. Try these things out for yourself and draw your own conclusions. (And please, let us know what you experienced.)

 

Hope this helps.

 

Best regards,

Barry

Soundkeeper Recordings

Barry Diament Audio

[Encore]

However, my experience has been that unless the ceiling is unusually low (or the speakers unusually high), this won't have anywhere near the effect that wall reflections do.

 

Why is this? The ceiling is of course (typically) further from the speakers than the floor, but not necessarily further from the speakers than the side walls ...

[bdiament]

Hi Encore,

 

Why is this? The ceiling is of course (typically) further from the speakers than the floor, but not necessarily further from the speakers than the side walls ...

 

It depends on the shape of your room and how your system is oriented within it.

Of course, there are situations when the ceiling will be closer than the side walls, this is true.

But most speakers in my experience will send more information to the sides than they will send upward.

 

Please keep the context of what I said:

"Ideally, I'd want to treat the ceiling reflections too..."

 

Best regards,

Barry

Soundkeeper Recordings

Barry Diament Audio

[Kiwi2]

The manufacturer would usually spec the response with and without the plugs.

 

No they don't. It is room dependent as to what effect the plugs will have.

 

 

Just my opinion of course but I would never assess "best response" for listening any other way than by listening.

 

More than likely you would prefer the configuration they gave the smoothest frequency response free of large peaks and dips. Maybe you could hear and notice 20 dB swings? If it sounds better without a large 60hz 10 dB FR peak (which would also be causing a long slow decay) then how about fine tuning out 5 dB peaks or dips?

 

And I am not talking about using EQ. Merely fine tuning speaker placement in the room.

[spdif-usb]

By the acoustic dimensions of spaces. Once boundaries start deviating from the ideally stiff and non-leaky norm the acoustic dimensions too deviate from the measured dimensions. This explains why the deepest room modes sometimes are off their pre-calculated values. Idem for the exact locations of nodes.

Yes, exactly. The point I was trying to make in my previous reply is that the absorptive character of wall coverings etcetera changes the frequency response of reflected sounds, and the stiffness of the walls changes the exact locations of nodes so, therefore, measurements are most often very useful indeed (that is, provided that measurements are sufficiently accurate in both the frequency domain and the time domain).

[bdiament]

Hi Kiwi2,

 

No they don't. It is room dependent as to what effect the plugs will have.

 

Perhaps our experience is different. Manufacturers I've dealt with do spec it and the arrangement which is the least peaky is not at all room dependent, it is design dependent. The *degree* of peakiness will depend on the room and on placement but not plugs vs. no plugs. Generally the plug will be less efficient but also flatter in response. It is basically a trade between quality and quantity. There are adherents to both paths.

 

 

More than likely you would prefer the configuration they gave the smoothest frequency response free of large peaks and dips. Maybe you could hear and notice 20 dB swings?...

 

I would certainly hope so, especially since applying EQ during mastering often involves fine tuning at fractions of a dB.

 

Best regards,

Barry

Soundkeeper Recordings

Barry Diament Audio

[bdiament]

Hi spdif-usb,

 

Yes, exactly. The point I was trying to make in my previous reply is that the absorptive character of wall coverings etcetera changes the frequency response of reflected sounds, and the stiffness of the walls changes the exact locations of nodes so, therefore, measurements are most often very useful indeed (that is, provided that measurements are sufficiently accurate in both the frequency domain and the time domain).

 

What I've found is that the materials and construction will change the *frequencies* of the resonant modes, not their physical locations and hence not the locations where they can be successfully treated. Pressure is always greatest at the boundaries, particularly where boundaries meet. Harmonics of the fundamental resonant modes occur at fractional lengths along these dimensions (first harmonic at 1/2 point, second harmonic at 1/4 points).

 

Maybe I just haven't experienced a room such as you describe. What I said above has been true in every room I've worked in for the past several decades, including domestic and studio situations where materials and dimensions have never been the same twice.

 

Best regards,

Barry

Soundkeeper Recordings

Barry Diament Audio

[spdif-usb]

What I've found is that the materials and construction will change the *frequencies* of the resonant modes, not their physical locations and hence not the locations where they can be successfully treated.

Low frequency sound waves can pass through room boundaries if those room boundaries aren't completely rigid (e.g., a wall made out of gypsum board) and, as Fokus already said, the room dimensions are also effectively changed if the air pressure of those sound waves causes air to escape through gaps (e.g., underneath a door, through a chimney connected to a fireplace,...).

 

On top of this, room modes are not necessarily easy to calculate even regardless of whether both conditions are met. Pieces of furniture placed on the floor can be thought of as irregularities in the shape of the floor, etcetera.

[bdiament]

Hi spdif-usb,

 

Low frequency sound waves can pass through room boundaries if those room boundaries aren't completely rigid (e.g., a wall made out of gypsum board) and, as Fokus already said, the room dimensions are also effectively changed if the air pressure of those sound waves causes air to escape through gaps (e.g., underneath a door, through a chimney connected to a fireplace,...).

 

On top of this, room modes are not necessarily easy to calculate even regardless of whether both conditions are met. Pieces of furniture placed on the floor can be thought of as irregularities in the shape of the floor, etcetera.

 

I hear what you're saying but no room I've ever set up, domestic or studio, regardless of construction, over several decades now, has ever departed from what I described. It doesn't matter if the walls are gypsum or cinder block, I've heard the same approach work in every room in which I've tried it.

 

I've never had a need to calculate room modes. I just go after them and get rid of the problem, using the same solution in every room I've ever needed to. Again, it has worked for me every time.

 

I don't care what frequencies the modes are at; I just want to change their Q so they are no longer an issue. Using mechanical means to do this has worked for me without exception.

 

Perhaps you've had different experiences (assuming you've tried the methods I describe - which are by the way, not my own invention but things I learned from the writings of Peter Walker and Harry Olsen, which date back over half a century and Art Noxon). I am reporting what has worked for me consistently for a long time in many, many rooms. It has proven itself over and over again to be fast, easy and effective.

 

If you've got a different approach that works for you, I think that is great. I'd be curious to hear about it and in how many spaces you've implemented it to your satisfaction.

 

Best regards,

Barry

Soundkeeper Recordings

Barry Diament Audio

[spdif-usb]

If you've got a different approach that works for you, I think that is great. I'd be curious to hear about it and in how many spaces you've implemented it to your satisfaction.

I agree the best way to treat a room is to position the speakers and the listening chair in such way that peaks and nulls become least intense overall in the part of the spectrum that's most difficult to tame using acoustic panels (i.e., the entire bass region of the spectrum, although paying gradually less attention to frequencies the further they are below 30 Hz) and work from there. Small rooms and irregular shaped rooms tend to be much more difficult to deal with, so the basic rules of thumb are not always applicable, nor always practical, and tradeoffs have to be made. Narrow dips and narrow peaks in the frequency response curve can be misleading, which is why looking at a waterfall plot with octave smoothing usually comes in handy, as it helps to identify the root cause of the problem.

 

The more complex a room gets, without accurate measurements, the more it all becomes a game of chance when trying to obtain good bass sounds. Sometimes, adding parametric EQ might yield better results than desperately trying to avoid using it. Other times, it might not. Unless perhaps you have the luxury of the near-perfect room, IMO there is no standard answer to which approach works best.

 

P.S. - Hiring Jeff Meier always works. lol

[Hipper]

My experience, with my speakers of course, is that ceiling reflections, if there were any, made no difference to the sound, but that side wall reflections did. However these sides wall reflections had a positive effect on the sound, making it a bit more lively. This is what Floyd Toole suggests.

 

My speakers have about a 30 degree toe in and it is said the mid and high drivers have a 30 degree dispersion but clearly there is still some sound beyond that.

[mayhem13]

My experience, with my speakers of course, is that ceiling reflections, if there were any, made no difference to the sound, but that side wall reflections did. However these sides wall reflections had a positive effect on the sound, making it a bit more lively. This is what Floyd Toole suggests.

 

My speakers have about a 30 degree toe in and it is said the mid and high drivers have a 30 degree dispersion but clearly there is still some sound beyond that.

 

Dispersion isn't going to be even through the frequency response. Higher frequencies through each driver begin to 'beam' on axis as the wavelength the produce meets or is less than that of the drivers dimensions.

 

As to the ceiling, you can be absolutely certain it has an effect on what you hear. One only needs to imagine the room without a ceiling to know it will sound drastically different.

[Hipper]

I put sonic insulation on the ceiling at the appropriate positions, and later removed it.

 

Whilst I couldn't guarantee by measurements that there was no difference, I could not detect any in everyday listening.

 

In comparison, I could detect side wall reflections after doing the same thing.

[Jud]

Dispersion isn't going to be even through the frequency response. Higher frequencies through each driver begin to 'beam' on axis as the wavelength the produce meets or is less than that of the drivers dimensions.

 

As to the ceiling, you can be absolutely certain it has an effect on what you hear. One only needs to imagine the room without a ceiling to know it will sound drastically different.

 

Interestingly, I've done this "experiment" in real life. I lived in a restored one-room schoolhouse originally built in 1869 (red brick, white bell tower with functioning bell...) where my listening room was over 30 feet high. Then we had a home built in 2004 where my listening room is 9 feet high. The system was the same, including the same speakers I still have, and the relationships in the "footprint" were and are very similar (distance between speakers, from speakers to listening position, from speakers to back wall, toe-in, etc.). There was a bit of difference in perceived soundstage height (higher in the higher room), but I must say not nearly as much as I'd feared.

[Blake]

Jud: Thanks for posting that experience. Mayhem is correct about ceiling reflections. However, I have always wondered about the real-world implications/audibility of the impact of ceiling reflections but I have never had to opportunity to check this out for myself.

[Jud]

Jud: Thanks for posting that experience. Mayhem is correct about ceiling reflections. However, I have always wondered about the real-world implications/audibility of the impact of ceiling reflections but I have never had to opportunity to check this out for myself.

 

I'm sure my experience would be relevant for everyone else living in 130-year-old timber frame brick buildings with foot-thick walls, 35-foot ceilings, half-floor open loft bedrooms, and Vandersteens.

[mayhem13]

Interestingly, I've done this "experiment" in real life. I lived in a restored one-room schoolhouse originally built in 1869 (red brick, white bell tower with functioning bell...) where my listening room was over 30 feet high. Then we had a home built in 2004 where my listening room is 9 feet high. The system was the same, including the same speakers I still have, and the relationships in the "footprint" were and are very similar (distance between speakers, from speakers to listening position, from speakers to back wall, toe-in, etc.). There was a bit of difference in perceived soundstage height (higher in the higher room), but I must say not nearly as much as I'd feared.

 

.....and I wouldn't expect there to be much change in perceived soundstage height as the key signals for localization in the 1-3khz range are pretty much beaming their spacial cues to the listener. What is drastically different is the acoustics of the space overall, with a clearly audible difference in tonality and coloration.

 

Of course, raising the ceilings to such heights helps with ceiling bounce, particularly with bass and if I had my choice, I'd have a loft type space with 20ft ceilings myself. It sounds like a great listening space. That being said, even extreme distances aren't immune to smearing reflections, and in some cases so objectionable that they're unacceptable. As an example, a live venue I used to house engineer annually was a Reggea festival in Coney Island that, due to the proximity of 500 yards or so to some large buildings, created a slap back delay that literally ruined the performance for anyone not within 50 yards of the stage. I'm sure others have experienced these same artifacts in large enclosed arenas or concert halls....or simply calling out in a canyon or such. Large spaces are great, but can have their own, sometimes more difficult to handle issues.

[Encore]

But most speakers in my experience will send more information to the sides than they will send upward.

 

So their vertical dispersion is less than their horizontal dispersion? But doesn't the same ideal of having good dispersion apply in the vertical dimension as well? Or put another way, if the ceiling becomes less critical because loudspeakers don't disperse as much vertically, wouldn't it likewise be a good way of diminishing the effects of wall reflections to have loudspeakers with less horizontal dispersion?

 

 

Please keep the context of what I said:

"Ideally, I'd want to treat the ceiling reflections too..."

 

I do ;-) I was just curious about the reasoning for the ceiling being less problematic.

[bdiament]

Hi Encore,

 

So their vertical dispersion is less than their horizontal dispersion? But doesn't the same ideal of having good dispersion apply in the vertical dimension as well? Or put another way, if the ceiling becomes less critical because loudspeakers don't disperse as much vertically, wouldn't it likewise be a good way of diminishing the effects of wall reflections to have loudspeakers with less horizontal dispersion?...

 

I think good arguments can be made for "controlled dispersion".

Certainly the speakers I've come to like the most tend to have reduced output to the sides, not just vertically.

 

Still, as I suggested earlier, the direct sound from the speaker is not the only sound to engage the room. The room interacts with *all* sounds occurring within it, so I would deem the ultimate treatment to cover all the boundaries. It just so happens that in many rooms (certainly not all) leaving out the ceiling treatment will not cause a great deal of damage. But of course, that is not the same as no damage at all. ;-}

 

Best regards,

Barry

Soundkeeper Recordings

Barry Diament Audio