diyAudio (
-   Planars & Exotics (
-   -   Panel speaker placement & room coupling (

fivestring 10th August 2010 01:27 PM

Panel speaker placement & room coupling
Hello all,

with permission from the author, Ivan Li from Hong Kong, I would like to start a discussion regarding the optimum placement of panel, dipole type speakers. His main speakers are Maggies, but I believe same goes for any dipole type, even coned speaker.
The optimum position may appear a bit extreme at the first glance, but lets discuss what`s best from the sound quality point only and leave out practical considerations for a moment.


Few of us, except perhaps the rare species of odd fossils of the pre-historic era, would have questioned the paramount importance of speaker placement nowadays. What is more debatable is where and how the speaker positions are to be determined. I have been a faithful follower of the room-coupling school and I believe that should be the best way out.

Room coupling, I am quite positive by now, really goes beyond the deliberations over direct verses reflected sound sources, because the focal point remains relatively constant irrespective of what damping treatments we have on the wall surfaces in any given room. This focal point does not seem to move until the dimension of the room is drastically changed.

To achieve effective room coupling, one has to locate the focal spot of the room first. There is a simple yet effective method to find this spot but I shall come back to that a while later. Every room has its own peaks and valleys acoustically. The peaks are resonances and the valleys are just the opposite, the suck-outs as a result of phase cancellation. I used to believe the coupling point is the spot where multiple resonances gravitate since it clocks the highest sound level throughout the audio band. If we look at it the other way round, however, the focal point of the room should more correctly be taken as the spot where phase cancellation is the lowest.

When the speakers are placed along this focal spot, phase cancellation would be at its minimal obtainable within the room and the system will then be able to resolve, with the best of fidelity and the least of distortion, the myriads of spatial information contained in the recording. The width and depth, the ambience, the layering and instrument placements etc. will all spring to life. Proper sound staging, no doubt, has much to do with this phase coherency across the audible frequency range. We all know what would happen if one speaker is inverted in phase—there will be no imaging, no soundstage, and even no sense of direction. This is the result of serious phase cancellation. Even if the speakers are properly in phase, however, there would still be a fair amount of cancellation depending on where the speakers are placed.

Ideally speaking, the room should be symmetrical in overall shape and the speakers symmetrically placed along the lengths of the rectangle. If one speaker is close to the side wall while the other is in the middle of the room, a rather common sight given the popular L-shape layout of sitting rooms in HK, there exist more chances for cancellation taking place at various frequencies, rendering the imaging blurred, stage collapsed and ambience lost.

Once the speakers are coupled to the room, the two merge into one. The room becomes an effective extension of the speakers which in turn would cease to exist visually. Tuning for solid imaging then becomes much easier. Human ears locate the sound source by detecting the time difference of direct sound arriving at the ears. To achieve three dimensional imaging, all we have to do is to cut down secondary reflections from overwhelming the direct sound. While on this subject, I like to point out that I have tried the live-end-dead-end approach and it did not work to my satisfaction. I believe speakers are designed with the average western style living room in mind where upholstery, curtain, carpet and furniture all contribute to an acoustic environment which is neither too dead nor too live, but moderate throughout. This is something worth considering when we set about fabricating our listening area. It is not uncommon to find that the more elaborately contrived the HiFi room is, the less satisfactory it often turns out.

All in all, phase coherence does appear to be the key and with this key we stand every good chance to unleash a completely new dimension of audio realism.


Here comes the core—the action part of the whole thing, a method that has proven to work wonders for me over the last 30 odd years.

Have the speakers placed along the lengths of the room, about Ľ to 1/3 from the back wall, and Ľ the breadth roughly. Then play some vocal music, the hilarious type, the more instruments the better.

Now walk slowly to and fro along the mid-line between the speakers, from one end of the room to the other and then back, may be several times to get the mind set (if the speakers are standing low, you may well have to crawl). Somewhere along the aisle, you would hear the sound getting louder and at the same time it rises above your head, filling the ceiling as if you’ve entered a Gothic church. Bingo, that is it, the focal point.

Mark this focal point and drag your speakers over the lateral line crossing the spot. You’ve just coupled your speakers to your room. What is left, may be the crucial part yet, is to find the perfect stereo seat. As the name suggests, you’ll have to look for the widest stereo soundstage again along the aisle, between the speakers. If one side of the room does not give you the best definition and soundstage, try the other side. I was the one having my seat moved over to the wrong side of the room to get the best out of my present system.

Final note, if you are unable to find the focal spot despite trekking up and down diligently the whole afternoon, your room dimension is probably too bad to be a listening room. Try another room if you have one. If not you’ll have to move house.

Ivan Li (Limage)

bentoronto 10th August 2010 06:25 PM

Ummm, the basic premise is false.

Each room mode has different "focal points." There may be one spot where ONE of the room modes (and some of its overtones) meet. But that's it.

Second, while a grand theory, my general POV (besides being allergic to grand theories) is that in something as real-world as a music room, you want to split the difference, straddle the mean, spread the effects, and so on... not fix things with one grand fix. Doesn't sound satisfying, but any examination of the research literature on spreading around a bunch of your subs hither and yon in disorder (in order to combat room modes best) will convince you otherwise.

Perhaps it is instructive that you and ESL manufacturers and owners think a kind of 1/3 location makes sense: that is likely to be the point in a room where you are straddling more modes and unstraddling more modes and spicing the sound with some rear-wave sound than anywhere else.

fivestring 10th August 2010 07:32 PM

OK, so you feel this method isn`t worth investigating because it lacks "scientific" evidence and approach? Fine, for all the others, another experience from someone that has tried it instead of only intellectualized about it, here.

Mitch Alsup 13th August 2010 06:53 PM


Originally Posted by fivestring (
Final note, if you are unable to find the focal spot despite trekking up and down diligently the whole afternoon, your room dimension is probably too bad to be a listening room. Try another room if you have one. If not you’ll have to move house.

I am not quite sure I buy the grand theory point you have, but some minor issues seam reasonable.

When I bought my CLS As in 1985-ish I had a room that was 24*16 with a semi cathedral ceiling. The speakers sounded just great and were not very sensitive to the room placement. However, the soundstage width was easily manipulated by toe-in of the speakers.

In 1992 I moved to a new house with a real cathedral ceiling and a room 26*24 and I could never get the system to sound correct. I tried a bunch of things but it just never worked.

In 2007 I moved into a house with what is now called the great room {Living room, dining room, and kitchen in one long room. This thing is 17 wide, 10 tall and 53 feet deep and sounds magnificent. In this room the CLS (now in Zii form) are only a little picky about room placement. They are very sensitive to how far from the lateral wall and they are sensitive to whether the dipole is pointed at the corner. So, seting these up for this room is an optimization problem where the latteral and the dipole can combine with the speaker toe so that the soundstage is properly wide and the image focused. The speakers want to be within about 1 inch of a certain lateral placement (22") and the dipole wants to be pointed 5" laterally out from the corner in order to sound magnificent.

I also added a Velodyne DD-15 subwoofer and used its control to integrate with the CLSs and dampen the room. The TV image shows near perfect response from 20Hz through about 200 Hz (where the DD setup program ends.) After installing the sub I played around with speaker placement again and basically came back to their original positions.

The other interesting thing about this room is that while the (THE) chair is just perfect for listening, the whole back half of the great room has near perfect acoustics where one can carry on a conversation while listening to medium loud music with narry a hit of directionality, dropouts, or room nodes. It is an extraordinary room for listening.

The placement I found optimal for these speakers does not quite fit into the grand theory you posit.

bentoronto 13th August 2010 07:08 PM

Nobody should proceed on the path of quality music reproduction without reading Toole.

Basic distinction between over-300 Hz and below situation in rooms. Funny thing, I've listened to my PopScience test record with the 300-20 glide tone of roughly 1958 hundreds in not thousands of times in the years since. Nice to have a yardstick. As Mitch says, the bigger the room the smoother the net room frequency compass down there. MANY "focal points" and those of us with a single sweet spot to serve are lucky.

Of course Toole's great 2008, 550 pp book is not for fivestring since he is very scientific and is "impatient"* with fivestring's POV. BTW fivestring, if you are raising a topic which you feel is not subject to rational (AKA scientific) analysis, why are you interested in anybody else's opinion?

*"Impatient".... took me a while to find a synonym for ".....". Wonderful living in Canada where the struggle for greater civility of expression has an earnestness even among erstwhile New Yorkers.

a.wayne 13th August 2010 09:11 PM

I would take dipole panel speaker setups with 1 big spoon and a ton of salt..


One size will never fit all and they are absolutely a nightmare to setup correctly and are affected by everything you could think of in your room...

bentoronto 13th August 2010 09:15 PM


Originally Posted by a.wayne (
I would take dipole panel speaker setups with 1 big spoon and a ton of salt..


One size will never fit all and they are absolutely a nightmare to setup correctly and are affected by everything you could think of in your room...

... and that differs from every other loudspeaker in what way?

a.wayne 13th August 2010 09:30 PM

Do you mean compared to monopole box speakers ? ..... a lot...

fivestring 14th August 2010 08:47 AM

Mitch, I`m not sure I understand how far from the back wall are CLSes and what is the XO frequency to the sub?
By the way, this is not my theory, I think I have stated that clearly, but it does reflect my personal experience that dipoles need to be placed really far away from the back wall and that close proximity to side walls helps in low frequency extension and overall quality.
At least two feet is needed to get any useful bass though at this position the early reflections from the back wave are way to annoying, placed three feet from the back wall ameliorates things a bit but not much, still too much reflections that cause phasey, confused soundstage and colorations in the upper ranges. Beyond five to seven feet things start to get very interesting and by the time I reach ten feet, I get the impression I`m listening to a totally different loudspeaker, everything is on a completely new level. Of course, with speakers positioned almost in the middle of the room, now I have the opportunity to listen to them from both sides, which gives interesting sound perspectives (my room is 22 x 11 feet). The measurement shows the frequency response at the listening position (speakers placed 10 feet from the back wall), two meters away from the speakers, smoothing was 1/12 per octave - looks really bad, huh? The room gain and the fundamental of the room is clearly visible (25 Hz), subjectively it sounds extended (22 Hz could be clearly heard and felt) and extremely well balanced and transparent through the whole spectrum.

bentoronto, I`m not that scientific at all, just as much as needed and when needed and I do have Toole`s book, I just find most of those great authors` findings useless in a sense that I still need to hear a dome/cone speaker that sounds remotely as transparent and natural as any planar, OB, ribbon, planar magnetic or ESL combo.

There are testimonials on the net of the ML Vista owners that claim they cannot imagine going back to cones`n`domes again after listening to this hybrid, the difference in sound is so big that only the deaf or those that have been exposed to dome torture for years and have become accustomed to their annoying sound, accepting it as a norm, could overhear it.

bentoronto, I can imagine it must be a heavy burden to know it all, be an ultimate expert and so smart that this prevents you from trying certain methods that don`t come from the "trusted" sources, well, the loss might be on your side.

All the best,

bentoronto 14th August 2010 09:32 AM

I'm laughing out loud... 11x22 room, you say. Do you even know why you should be blushing?

You really need to read Toole, think twice before telling the world you can hear the difference between 22 and 25 Hz, say how much smarter you are than the whole profession of acoustics, or heap gratuitous personal (not content-based) insults and put-downs on people you don't know who respond to your comments.

I'm outta here.

All times are GMT. The time now is 01:07 PM.

Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 18.75%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Copyright ©1999-2018 diyAudio