Optimum dispersion in speakers

[gshelley]

Really enjoying this thread and learning more on speaker types& room Acoustics.

Duke,
please elaborate “ (I take a controversial position regarding early sidewall reflections - I believe they are generally overall undesirable, though they do have desirable aspects. I can go into detail if anyone would like.)

and “implications of dispersion on sweet spot width... some of which is intuitive, some of which is not.”
Many thanks!

 

[bonzo75]

actually an additional reason I like horns is I don't need to sit in the sweet spot. You can move around the room and if you have an open plan kitchen, you can cook, eat, lounge, all while listening to music..

As mentioned one of my favorite rooms is Silvercore's, which is almost 60 feet long and over 20 feet wide, plus an L at the far end (it is a warehouse). You can listen to music 3 feet away, 10 feet away on the sofa, and 50 away on the dining table. Additionally he has the WE 16a which you can listen to at almost 180 degrees.

Do some horns sound better in the sweet spot as compared to outside it? Yes. But with pnoe, universum, Altec with multi cell dispersion, you don't really need to be in the sweet spot. Unless you have tractrix then yes

 

[Folsom]

actually an additional reason I like horns is I don't need to sit in the sweet spot. You can move around the room and if you have an open plan kitchen, you can cook, eat, lounge, all while listening to music..

As mentioned one of my favorite rooms is Silvercore's, which is almost 60 feet long and over 20 feet wide, plus an L at the far end (it is a warehouse). You can listen to music 3 feet away, 10 feet away on the sofa, and 50 away on the dining table. Additionally he has the WE 16a which you can listen to at almost 180 degrees.

Do some horns sound better in the sweet spot as compared to outside it? Yes. But with pnoe, universum, Altec with multi cell dispersion, you don't really need to be in the sweet spot. Unless you have tractrix then yes

I actually do a majority of listening out of the sweet spot. I don’t believe that’s entirely a horn only thing, but certainly not much of a normal box speaker out into the room thing.

 

[Duke LeJeune]

Duke,
please elaborate “ (I take a controversial position regarding early sidewall reflections - I believe they are generally overall undesirable, though they do have desirable aspects. I can go into detail if anyone would like.)

Early same-side-wall reflections increase the "apparent source width", to use Floyd Toole's terminology. In other words, they can made the soundstage sound like it extends laterally beyond the left and right speakers (even if such information is not on the recording), which most listeners (including yours truly) think is pretty cool.

But this can come at a price.

Quoting Earl Geddes on the subject: "The earlier and the greater in level the first room reflections are, the worse they are. This aspect of sound perception is controversial. Some believe that all reflections are good because they increase the listeners feeling of space – they increase the spaciousness of the sound. While it is certainly true that all reflections add to spaciousness, the very early ones (< 10 ms.) do so at the sake of imaging and coloration."

The same effect which spreads the image wider also smears the precise location of sound images. And regarding coloration, what happens is this: Once the direct sound and that early same-side-wall sidewall reflection have mixed in the ear canal, they are combined. The most beneficial early reflections are the ones which arrive at the opposite ear from the direct sound, because the ear/brain system can process them separately.

Here is Earl Geddes' paper on his speaker design philosophy, most of which I subscribe to. See page five in particular:

http://www.gedlee.com/Papers/Philosophy.pdf

Also, I think that strong early same-side-wall reflections can contribute to a "small room signature" which tends to dominate over the (highly desirable) spatial cues on the recording. (Imo we want to minimize "small room signature" and effectively present the recording's acoustic signature, whether it be real or engineered.)

To offset the loss of soundstage width from minimizing those early same-side-wall reflections we can move the speakers a bit further apart, but I must admit that "soundstage wider than the speakers" thing was pretty cool.

Toole finds that listeners generally prefer good wide-pattern speakers over good narrow-pattern speakers, but I'm not convinced that it is because those early same-side-wall reflections are desirable. I think it may be because having more reverberant energy is generally desirable, and that more than offsets any detrimental effects of those early same-side-wall reflections. (The approach I use results in approximately the same amount of in-room reverberant energy as a good wide-pattern speaker but without significant early same-side-wall reflections, so it's like I'm taking Toole's wide pattern and chopping it up into two narrower slices and serving them up a separately.)

and “implications of dispersion on sweet spot width... some of which is intuitive, some of which is not.”

The ear localizes sound by two mechanisms: Arrival time and intensity. (Most image localization happens north of 500 Hz.)

When you are seated in the central sweet spot, arrival time is the same from both speakers, and so is intensity. This is the ideal. But what happens when you move off to one side?

Well if the speakers have a very narrow radiation pattern (like a wide, flat-panel eletrostat) and are toed in just enough to cover the sweet spot, when you move off to one side a bit you may well be moving outside the effective coverage pattern of the far speaker. So now the near speaker "wins" arrival time slightly, but it "wins" intensity by a very large margin. So the sound images are pulled hard towards the near speaker.

Suppose we are listening to conventional wide-pattern speakers, with some toe-in. Once again when we move off to the side a bit the near speaker "wins" arrival time, and it "wins" intensity as well, largely because we are more on-axis of the near speaker and further off-axis of the far speaker. So the image is still pulled towards the near speaker, but not as much.

Next let's try omnidirectional speakers. Now we are equally "on axis" of both speakers even when we are off the centerline, so the intensity discrepancy is reduced, but the image is stilled pulled somewhat towards the near speaker.

We can significantly reduce the intensity discrepancy by using speakers which approximate a line source, as SPL falls off by only 3 dB for each doubling of distance from a line source, in contrast with 6 dB per doubling of distance from a point source. Sound Lab's big faceted-curved panels have wide enough patterns that you can move off to the side a bit and still be well within the coverage pattern of the far speaker. So the near speaker (which ineviably "wins" arrival time) still "wins" intensity, but by a very small margin, so the image is only pulled a little bit.

Now for the counter-intuitive part: Suppose we take a pair of speakers with constant-directivity horns whose pattern is about 90 degrees wide (-6 dB at 45 degrees off-axis to either side). And let's toe them in severely, like 45 degrees, such that their axes criss-cross in front of central sweet spot. Something very interesting happens as we move off to the side:

Off to the side of the centerline, we are now quite far off-axis of the near speaker, but on-axis (or nearly so) of the far speaker! So the near speaker inevitably "wins" arrival time, but the far speaker "wins" intensity! The result is that we still have a pretty good spread to the instruments even from well off to either side. This doesn't work perfectly, and some off-centerline locations are better than others, but the net result can be enjoyable even from outside the plane of the speakers!

At audio shows I usually set one chair up against a side wall, outside the plane of the speakers, and every now and then the room is so full that somebody reluctantly sits in it. After the song I will ask that person how it sounded from there, and they are invariably pleasantly surprised. I ask if they could still hear an enjoyable spread of the instruments and they have always said yes.

The SECRET to this extreme toe-in technique working well is, the output of that near speaker must fall off smoothly and rapidly as you move off-axis. So it calls for a rather unusual set of characteristics, which most speakers do not have.

Thus, in my opinion (and somewhat counter-intuitively), the widest sweet spot comes from a specific type of fairly narrow-pattern speaker, set up a certain way.

And do you recall just above when I said that "the most beneficial early reflections are the ones which arrive at the opposite ear from the direct sound, because the ear/brain system can process them separately"? Well keep picturing that aggressive 45-degree toe-in we used with our horns to get that wide sweet spot: The left-side speaker is toed in so much that its pattern "misses" the left-side wall, but instead it is aimed across the room towards the right-side wall. So the first strong lateral reflection for the left-side speaker will be off the OPPOSITE wall and will arrive at the OPPOSITE ear, and same for the right-side speaker! Wide sweet spot + spaciousness without coloration, two birds with one stone. Credit to Earl Geddes for both of them.

 

[Hipper]

Regarding early side wall reflections, I've read (in Toole) that these reflections, if very similar in frequencies to the direct sound and arriving within a short time (say 10ms) add to the direct sound. That is they only make it louder due to the so called Haas effect and the sum is perceived as a single sound. There are no directional cues. How then can such early reflections offer spatial cues?

My limited home experience in a small room (14' x 13' x 8' high) is that side wall reflections muddy the sound and offer no obvious spatial cues. Removing them gives a much 'crisper' sound. I should add though that my musical tastes are mostly studio bound pop.

I would guess that the Haas effect (or precedence effect) is a personal experience that varies with individuals and frequencies. Is that so?

On using speakers as giant headphones, I've often wondered about that but never tried. It seems to me that above the transition frequency of a room (say 300Hz), and with good absorption, you would receive the left speaker sounds only in the left ear etc., so avoiding interaural crosstalk which some say is beneficial (ambiophonics). Where it may not work is below the transition frequency, although I would think with a wide enough room and bass traps or DSP these problems could be dealt with.

 

[Duke LeJeune]

Regarding early side wall reflections, I've read (in Toole) that these reflections, if very similar in frequencies to the direct sound and arriving within a short time (say 10ms) add to the direct sound. That is they only make it louder due to the so called Haas effect and the sum is perceived as a single sound. There are no directional cues. How then can such early reflections offer spatial cues?

My understanding is that the two are perceived as a single sound but that the apparent location can be pulled to the side a little if the reflection is loud enough that the Haas effect isn't 100% effective. However I may be mistaken; it has been a long time since I read about it.

My limited home experience in a small room (14' x 13' x 8' high) is that side wall reflections muddy the sound and offer no obvious spatial cues. Removing them gives a much 'crisper' sound.

That improvement in clarity from removing the early sidewall reflections matches my experience as well. The last time I claimed that happened resulted in a very long and inconclusive (but amicable) exchange on another thread.

Researcher David Griesinger, whose work is focused on concert halls and large room acoustics, has some interesting things to say which imo are relevant to small rooms as well (he doesn't use qualifiers which would exclude small rooms, and to me these statements seem to be applicable):

"The earlier a reflection arrives the more it contributes to masking the direct sound."

"If you want to communicate with sound, you need to make the direct sound distinctly audible.”

"Presence depends in the ability of the ear and brain to detect the direct sound a separate from the reflections that soon overwhelm it. When presence is lacking the earliest reflections are the most responsible."

From reading Griesinger, I think he uses the word "presence" the way I might use the word "clarity". The ideal would be when we have "presence" combined with "envelopment", which is what happens in a really good seat in a concert hall. Quoting Griesinger again:

"Envelopment is perceived when the ear and brain can detect TWO separate streams:

"A foreground stream of direct sound.

"And a background stream of reverberation.

"Both streams must be present if sound is perceived as enveloping."

Translating these principles to our much smaller listening rooms, we want to minimize early reflections (those arriving before 10 milliseconds, to use Geddes' criteria), while retaining plenty of later-arriving spectrally-correct reflections.

At least, this is my understanding.

 

[gshelley]

Thank you for the additional insight and the paper. Will spend some time reading it too.

 

[ferenc_k]

It is a very interesting subject.

When I was selling Danley speaker for PA, I tried some of them at home as well and ended selling my Avantgarde Trio being converted to a Danley user. The very tight dispersion of Danley can be beneficial not only at a PA distance but at home as well. If you have a room with some strong reverberation, you can try to put a Danley with 50x50 or so pattern into the corners of the room. This way (because of the very tight dispersion and strong pattern control of the Synergy horns) you practically will not have any early reflection as the speaker project the sound parallel of the wall. A bit on a way how a Klipschorn does this but much better. You can adjust the listening position accordingly then. This way these PA speakers can disappear from the sound and leave you a 3D sound field if the recording has any.

 

[Gregadd]

I wonder if we can define the "sweet spot.'
Best:
Freqiiency response
Best imaging and depth
Least room interference
Least distortion
Should it be wide or narrow
Some of which has been touched on already.

 

[ferenc_k]

I wonder if we can define the "sweet spot.'
Best:
Freqiiency response
Best imaging and depth
Least room interference
Least distortion
Should it be wide or narrow
Some of which has been touched on already.

I think, not the sweet spot itself is the most interesting. But a disappearing speaker. To build a setup, where you do not realize there is a speaker at all. The big if: IF the recording is done such a way where there is real, not reverbs generated space.

 

[Folsom]

I wonder if we can define the "sweet spot.'
Best:
Freqiiency response
Best imaging and depth
Least room interference
Least distortion
Should it be wide or narrow
Some of which has been touched on already.

Good power response throughout the frequency spectrum and low crosstalk per ear. That's it. All the other stuff you mentioned are a part of that.

Wide or narrow just depends on the type of speaker, so it's a personal choice.

 

[Gregadd]

How do we achieve low crosstalk?
Is it speaker design?Speaker placement?
Most good speakers now days can pull of the disappearing act to some degree. It does require some mental gymnastics to forget thier is a speaker there.
I do think we hear the same thing in both ears. It arrives at different times. Perhaps at at a different amplitude.

.

 

[Folsom]

How do we achieve low crosstalk?
Is it speaker design?Speaker placement?
Most good speakers now days can pull of the disappearing act to some degree. It does require some mental gymnastics to forget thier is a speaker there.
I do think we hear the same thing in both ears. It arrives at different times. Perhaps at at a different amplitude.

.

Crosstalk in this case is just not having bad reflections. Reflections from the other speaker, and even the one primary to the side of the ear it's on, can muddle sound if they're really bad. Directivity, placement, and room size are the first factors, then how you're aiming the speakers matters. If you have high directivity but they're aimed wonky you still get issues. IMO this is belittled by power response though.

While there's a lot of information surrounding the sweet spot, it's not too hard to achieve with most of our simple known rules.

 

[Gregadd]

Sheldon finds acoustic sweet spot.

 

[kach22i]

The general rule of dispersion is, the wider the radiating surface at a given frequency, the narrower the dispersion at that frequency (bending wave drivers and curved panels being an exceptions)......

Duke, what classification would Exciter loudspeakers fall under?

I called them "Activator" before, some call them surface vibration type.

 

[Duke LeJeune]

Duke, what classification would Exciter loudspeakers fall under?

I called them "Activator" before, some call them surface vibration type.

My understanding is that bending wave loudspeakers tend to have very wide dispersion, but this is not a type of speaker I know much about.

 

[kach22i]

Thanks, I feel like I learned something from your posts.

 

[Loheswaran]

This is a really good thread.

I have a pair of Yamaha Ns1000M speakers, and love them. Sadly, not only did the bass overpower my little hifi room, but the mid and treble were all over the places. You perhaps know that the mid unit is also a large tweeter. As such the dispersion is very wide. The problem is that early first and in some cases second reflections are happening too quickly and contributed to absolutely no image stability. The sound was a total mess. I've kept the NS1000M for another room.

I then got some Roksan Darius S1's which have a ribbon tweeter. These have a narrow dispersion pattern. All of a sudden I got image stability and the soundstage has completely opened up

 

[kach22i]

I once did a DIY speaker experiment before I knew about L-Pad attenuators for tweeters. The design was super primitive but effective, even my wife liked it.

I bent a coat hanger in such a way as to cup a golf ball. The ball was in front of the tweeter about 1-1/2" suspended and mounted to mounting plate.

The reason for doing this was to balance with the other lower range driver as the speaker was too bright.

The fun part was the balls were free to rotate in the armature via vibrations and sort of did a hypnotic dancing show.

My point is the higher frequencies were now prematurely dispersed or diffused and blended well with the mid and lower frequencies.

I've also done some whizzer experiments, and read the dispersion patterns are affected with certain frequencies.

I suspect other shapes do similar things.

Isn't one company still putting an "acoustic lens" to split or flatten the wave pattern on their tweeters?

 

[kach22i]

Speakers like this will have you scratching your head.

2012
Meyer Sound CAL column array: Is voice clarity possible in reverberant spaces?

There is no shortage of stadiums, houses of worship, and reverberant environments that could use more sonic clarity in their public address. It is easy to blame the loudspeakers but what often gets missed is the lively acoustical nature of these physical spaces.

With its beam steering and splitting capabilities, the Meyer Sound CAL column array loudspeaker can vertically direct a sound beam to aim directly at the audience members who need to hear it without exciting an acoustic space.