speakers immune from room effects. can this even be possible?

[oldric_naubhoff]

some time ago I discovered one great (IMO) speaker invention in the form of a CBT36 speaker. I didn't turn your attention to this speaker because I knew few people here dig deep enough into technicalities behind hi-fi design to even start a thread. but today I discovered some flicks at youtube that might prove very useful in showing the advantages of the technology better than any graph could do. but first some foreword.

why CBT? it stands for Constant Beamwidth Transducer. as the name suggest the speaker should produce a stable wave launch regardless of the surroundings it's placed in. or in other words it should not matter if the speaker is playing in anechoic or echoic conditions because its performance will be unaltered. why 36? the speaker is slightly bent, it's like a fraction of a circle. and the bend arc is 36 degrees. the speaker was invented by speaker design veteran Don Keele.

and now the flicks I was mentioning above:

- the speaker itself in action (please read the introduction in the movie): http://www.youtube.com/watch?v=McbuV1jSOs8

- some theory behind the design by Don Keele himself and more action: http://www.youtube.com/watch?v=ImSOzhI5JJA

just for comparison's sake I link to some other flicks:

- a very hi-fi system, video courtesy of a forum member: http://www.youtube.com/watch?v=Lcs_emvt9gA (regards Iceman :wave: )

- and "the second best speakers money can buy" as some want us to believe, just to show you it's not a problem of passive speakers, or active, but all quasi point source speakers placed in an untreated room suffer similar problem: http://www.youtube.com/watch?v=slcpDA7adhw

note excessive echo camera mic picks up in both cases, and also you can hear considerable deviations from flattish frequency response at some points in frequency response spectrum. of course if you were in that room and heard the speakers for yourself they wouldn't sound half as bad due to our brain doing a lot of filtering out of unnecessary sounds and flattening some freq deviations to make sounds sound as the brain remembers them they should sound (if you don't believe in what I just said make a simple experiment - record your hi-fi playing through a mic (with no echo cancellation), like on a photo camera and then listen to the recording. I'm sure the revelation will be totally different to what you're accustomed to).

if, however, someone is interested in how the technology works I recommend checking out those pdfs:

- CBT36 vs. B&W 801 Matrix - comparison of in-room measured performance. the significance of this exercise is that it shows how your average quasi point source speakers interact with a room (you could replace B&W speakers with any other similar design and the results won't vary much) and how CBT36 seems to be immune to effects of early reflections (freq response deviations) and how they manage to maintain virtually unaltered character regardless of the listening distance from the speaker or the listening height: http://www.audioartistry.com/brochures/B&W%20801%20vs.%20CBT36%20Ground-Plane%20Measurements%20v8.1.pdf

- this pdf here shows graphic representation of how different speaker designs behave in "free air" a.k.a. anechoic conditions (highly unlikely to ever occur in practice to most) and on ground plane (most common situation). those graphs should give you some food for thought, I hope:

http://www.audioartistry.com/Poster/Poster%20-%20Vertical%20Sound%20Field%20Simulations%20v10%20(Final).pdf

- similar to the one above but much more in depth; comparison of performance of different line-array speaker systems: http://www.audioartistry.com/Papers/CBT%20Paper6%20PerformanceRankingof%20Line%20Arrays.pdf

maybe some pros and cons for the ending:

1. pros:

- as already mentioned - immune from room effects, uniform soundfield over a wide array of listening distances and heights.

- nearly purely resistive 4 Ohms impedance curve with little electrical phase shift - meaning very easy load on the associated amps, but need to be rated into 4 Ohms load. as a side word; this feature came to me as a big surprise because the speakers use conventional (albeit custom made) voice coil drivers. since there is a voice coil in the signal path there has to show up inductive load on impedance curve at some stage. but the speakers are designed in such a way that there are resistors in series with the drivers. nothing to do with impedance linearisation, it's simply like some of the driver banks need to be attenuated somewhat with relation to the others and flat impedance curve comes as a bonus. OK, some may argue that because of this any potentially beneficial effect of amps damping factor onto drivers' movement control is wasted (the driver will "see" impedance from the resistor and not the amp), but since there are so many drivers working in tandem none of them has to work too hard on its own. excursion will always be minimal, even at high volume levels, hence the effect of amp's DF shouldn't be crucial here IMO.

- as they say: de gustibus non est disputandum, but IMO the speakers look just stunning! very interesting looking design. and this is for a speaker where form follows function.

- IMO rather affordable price of $2000 per pair...

2. cons:

- ... but the speakers so far are only available in the DIY version, AFAIK. which means you get all the necessary parts, including drivers and cabinetry, but you have to do the whole assembly and finishing yourself, or let somebody knowing do it for you. AFAIK the speakers will also be offered in fully assembled version but at much higher price.

- not just plug-and-play. what I mean by that is in order to get the speakers up-and-running you'll also need an active crossover and 4 channels of amplification (like 2 stereo amps or 4 monos). the speakers are meant to be working in active principle, there's no option of a passive version. because of the associated equipment requirement definitely not an option for minimalists.

- the speakers are spec'ed to reach down to 45Hz (despite small woofers used). that should be low enough for most but if you're one of the nether region lovers you'll need augmentation from subwoofers (and a 3-way active xover if you're a perfectionist ).

I'm hoping all that English would come useful to someone. please input below your thoughts on the whole idea, if you have any.

P.S. I'm thinking I'd like to see an integrated 4-channel amp with built in 3-way active xover on the market soon. That would solve the multi-box requirement to drive the speakers for me...

 

[steve_1979]

I'm no expert but this doesn't seem to make sense. No matter what the source of a sound is, if a sound wave is traveling through the air it's going to hit the boundaries of the room and bounce back as an echo.

Surely the source of the sound is immaterial. How is it possible for sound waves to be immune to room effects?

 

[MakkaPakka]

Also confused. I can't remember the exact figure but I think it's something like 60% of the sound you hear isn't direct from the speakers but from the room. Sound waves will bounce off whatever's close before they get to your ears.

 

[Craig M.]

I haven't checked, so might have this ar*e about face, but isn't the point of this that the sound is very directional? So it doesn't really come into contact with the floor, ceiling and side walls, apart from what hits the rear wall and then spreads around the room (at a much lower level)? I seem to remember reading something about a similar technology in the lead up to the London olympics, the plan was to use it as some form of crowd control with the ability to direct sound at an individual from quite a distance.

The demo video looked good, but the speakers form would rule them out for me. They would need too much space to be practical, but if they were much smaller the sound would possibly be too directional leading to a very small sweet spot.

 

[RobinKidderminster]

Simple wave theory would suggest that any loudspeaker radiates energy widely. Higher the frequency more absorbsion so tweeters are more 'directional' but this is a factor of frequency rather than tweeter/housing design. Maybe I too have missed the point but for me any soundwave, once 'released' into a room, will react with its environment irrespective of its source. Sound waves behave like ripples in water (except in 3D). On the other hand, does the shape of the speaker influence the interference patterns of waves? I feel the need to design a speaker system based on a parabolic enclosure maybe? Back to the drawing board .....
But then again. Maybe after further reading?? Independant test needed WHF. Now how DO you bend mdf?

 

[oldric_naubhoff]

I'm no expert but this doesn't seem to make sense. No matter what the source of a sound is, if a sound wave is traveling through the air it's going to hit the boundaries of the room and bounce back as an echo.

Surely the source of the sound is immaterial. How is it possible for sound waves to be immune to room effects?

it's not the sound waves that are to be immune to room effects in case of a CBT36, that's impossible. read on below.

Also confused. I can't remember the exact figure but I think it's something like 60% of the sound you hear isn't direct from the speakers but from the room. Sound waves will bounce off whatever's close before they get to your ears.

the idea behind the design is to make the speaker such that the sound reflections off the room boundaries has little, or no effect on the end effect for the listener. conventional speakers are hopefully designed using anechoic chambers (or at least measurements are taken and processed by software emulating anechoic environment). hence conventional speakers perform at optimal level in anechoic chambers. few, or even none of usual speaker users have the luxury to listen to their hi-fi in an anechoic chambers. usually it's the living room or some other room in the house. which means the performance of speakers is automatically compromised by listening in the environment they were not optimised to be listened to. of course, it's economically unfeasible to tune each and every speaker to every room. hence you get a one-size-fits-all product. and that's where CBT steps in. the technology on which it's principle it based on was developed and tested by US Army and declassified in the 80-ties (don't ask me why they would be working on something like CBT in the first place but that's what how it was). so what you get is a transducer that is capable to launch it's soundwave in such a way that combined in-room response doesn't differ much from anechoic response. furthermore, no room specific equalisation is necessary because the results will be similar in every room. I again refer you to this pdf:

http://www.audioartistry.com/Poster/Poster%20-%20Vertical%20Sound%20Field%20Simulations%20v10%20(Final).pdf

note how uniform power response of a CBT is compared to other designs. no lobes and no power response reinforcement regions. this is achieved by the special arrangement of the drives, their spacing and crossover point as well as shading between different sections of the array.

so basically what that means is in case of ordinary speakers room interaction has big influence on what you ultimately hear. in case of CBT the effect of room is negligible so you can hear more of what's on the recording.

I think this is as easy as I can explain. if still in doubt how this thing works like I recommend you read some white papers included on Don Keele's and Audio artistry's web sites. they will be more informative on the matter than I could ever be.

 

[oldric_naubhoff]

I haven't checked, so might have this ar*e about face, but isn't the point of this that the sound is very directional? So it doesn't really come into contact with the floor, ceiling and side walls, apart from what hits the rear wall and then spreads around the room (at a much lower level)? I seem to remember reading something about a similar technology in the lead up to the London olympics, the plan was to use it as some form of crowd control with the ability to direct sound at an individual from quite a distance.

The demo video looked good, but the speakers form would rule them out for me. They would need too much space to be practical, but if they were much smaller the sound would possibly be too directional leading to a very small sweet spot.

not in case of CBT. making speakers extremely directional is another way to skin the cat a.k.a. taking room effects out of the equation. but in this case you suffer very small sweet spot. in case of CBT you get very uniform soundfield over almost the entire room. you can freely walk around the room and the sound wont change considerably. if you check the graphs from the CBT vs. B&W knockout you'll note that the sound virtually doesn't change over nearly 150 degree horizontal arc and down from ground level up to about 2m. those are hardly directional speakers.

 

[RobinKidderminster]

Have you heard them oldric or are we simply discussing the literature? Guess always sceptical of unusual solutions but interesting never the less. My physics just doesn't understand how sound waves, once established, can behave differently from different sources with respect to boundry reflection/absorbsion.

 

[steve_1979]

...the idea behind the design is to make the speaker such that the sound reflections off the room boundaries has little, or no effect on the end effect for the listener...

You'll have to excuse my ignorance here. But I'm still confused as to how it works if it isn't just a very focused directionality speaker as Craig M suggested.

When sound waves hit a solid object (walls, ceiling, furniture etc) part of that wave will be reflected back. Could you describe to me exactly how the sound waves produced by this speaker are different to all other sound waves in this respect and why would the reflected sound waves have "little or no effect on the listener"?

If a sound wave doesn't reflect off a solid object then it must be disobeying well understood and documented physical laws of sound reflection. Am I missing something here or are the sound waves produced from this speaker different to all other sound waves in some way?

 

[oldric_naubhoff]

You'll have to excuse my ignorance here. But I'm still confused as to how it works if it isn't just a very focused directionality speaker as Craig M suggested.

directionality of this speaker depends on the arc. in case of 36 deg ground plane CBT you get 72 deg coverage in non reflective (save for the floor) environment. that's quite a wide beam IMO and not remotely close to what Craig was writing about.

When sound waves hit a solid object (walls, ceiling, furniture etc) part of that wave will be reflected back. Could you describe to me exactly how the sound waves produced by this speaker are different to all other sound waves in this respect and why would the reflected sound waves have "little or no effect on the listener"?

Steve, the main difference this type of speaker has over conventional design it produces very uniform sound field regardless of frequency. it doesn't produce any lobes in off axis frequency response. ill off-axis behavior of most speakers is mostly responsible for poorer or better subjective perception of speakers' sound.

you still get a lot of reflected sound even when using CBT. but the advantage is the reflected sound won't interfere as much with the directly projected sound. you won't get as many frequency response variations and deviations as in case of typical speakers.

I don't know if my post clears your doubt but maybe you'd like to hear what Don Keele has to say about his design. all parts shouldn't take you more than 1 hour:

http://www.youtube.com/watch?v=RaqVKy-7ruY

http://www.youtube.com/watch?v=aDdrlbCNBjE

http://www.youtube.com/watch?v=0ucfyAeug8E

http://www.youtube.com/watch?v=5UJtzpFWiNk

http://www.youtube.com/watch?v=alUVzX6bYOw

http://www.youtube.com/watch?v=MO8iAdoXnYc

If a sound wave doesn't reflect off a solid object then it must be disobeying well understood and documented physical laws of sound reflection. Am I missing something here or are the sound waves produced from this speaker different to all other sound waves in some way?

sorry to tell you that Stevie but you're obviously must be missing something here. the design is based on theoretically construed and then tested principle based on Lagendre equations. so no, the speakers don't disobey any laws. quite contrary IMO. what all that mathematical modelling contributed to is designing of a speaker which not only has flat frequency response on axis but also very much so off axis, or in other words the on axis region in front of the speaker is much larger than in case of your typical speaker and also there are no ill effects in frequency response off axis. BTW all those "ill effects" I was referring to mostly originate from suboptimal spacing between drivers and xover points used, in case of multiway speaker design. the influence of those "ill effects" will be further strengthened by close proximity of reflective surfaces (floor, ceiling, walls). such speakers should only be listened to in anechoic chambers on axis for best performance.

 

[oldric_naubhoff]

Higher the frequency more absorbsion so tweeters are more 'directional' but this is a factor of frequency rather than tweeter/housing design.

actually the beaming effect, you are referring to as "tweeters are more directional", is related to driver's width rather than absorptive properties of the listening room. of course how "live" or dead" the room is has its part too but this is a completely different story. as a rule of thumb; a driver will start beaming as soon as the wavelenght it produces becomes equal to the diaphragm width. so in case of a typical 1" tweeter dome it'll start beam frequencies from about 13.5 kHz upwards.

Now how DO you bend mdf?

in case of CBT36 you won't have to. I once read a thread about putting this thing together here. it turns out the front baffle is flexible and you just "lay" it over the rest of cabinetry when all driver assembly is over.

 

[oldric_naubhoff]

Have you heard them oldric or are we simply discussing the literature? Guess always sceptical of unusual solutions but interesting never the less.

never heard them and I'm afraid it's not going to be easy to get hold of a pair. unless you're planning to visit CES, or smth similar, you'll be hard pressed to listen to a pair. nevertheless, I learned that if something has strong theoretical backing to perform well and then measurements confirm the theory then the product is good regardless if you get to hear it or not. so basically I wish I could experience the speakers in action, but it won't be a deal breaker for me if I can't listen to a pair if I decided to buy them. the theory behind the design fully convinces me.

My physics just doesn't understand how sound waves, once established, can behave differently from different sources with respect to boundry reflection/absorbsion.

as you can read in other my posts in this thread I never claimed that. typical speakers measure very bad off axis due to unwanted interactions between drives due to bad driver spacing and bad xover points chosen. poor off axis performance contributes to poor perceived performance in a reflective listening environment.

 

[steve_1979]

Thanks oldric.

They look quite nice (to my eyes) and I'd like to have a chance to hear them for myself.

 

[RobinKidderminster]

Having suffered much of his lecture I concur ... an interesting development which will hopefully live up to its promise. Cheers.

 

[oldric_naubhoff]

Thanks oldric.

They look quite nice (to my eyes) and I'd like to have a chance to hear them for myself.

not a bother

I wish I could hear them too.

 

[oldric_naubhoff]

an interesting development which will hopefully live up to its promise.

I think it already did:

http://www.stereophile.com/content/don-keele-constant-directivity-crt36-speaker

(note "these speakers produced one of my best sounds at the 2011 RMAF." part)

and here too:

http://www.stereophile.com/content/audio-artistry-cbt36

(most relevant part "Most impressive to me, the sound remained consistent regardless of where in the room I sat or how close to the speakers I stood—a fun party trick").

you could of course find some more impressions on the web if you're interested.

 

[DocG]

Hey Oldric,

I had missed this thread completely (internetless holiday second half of August -- my therapist says it's good for me!).

I've now done some reading, mostly on the JBLs (I'm somewhat put off by the DIY aspect of the CBT36). Apart from the CBT100LA, there's also a CBT50LA and a CBT200LA, with half and double amount of drivers respectively. They have the same freq. response, the most obvious difference being lower/higher SPLs, but I notice some more differences. The vertical coverage is fixed at 20° in the 50 (I guess that won't be too much of a problem). But then they mention the 'Pattern Control Frequency', at 1500 and 600 Hz for the 50 and the 100 respectively. Does that mean the Constant Bandwith Technology only works properly from this freq. and up?

 

[lindsayt]

Yes, there's a lot to be said for line arrays when it comes to in-room dispersion. The curved line array approach of the CBT's looks interesting.

As well as good dispersion, line arrays also give you better power handling. If you use the right number and type of drivers wired in a combination of in series and in parallel you can also end up with gentle impedance characteristics. Another benefit is that the soundstage can have height, as opposed to dual concentric designs where the whole band or orchestra sound as if they are on a flat horizontal plane.

The downside of line arrays is the cost of the all those drivers, as well as the cabinets. Also for nearfield listening you get comb effects. And if you're using small drivers you need equalisation for extended bass, or additional large drivers.

Overall, I think there's a lot to be said for line arrays. And there's nothing new about them. They've been around since at least the 1960's.

 

[Andrew17321]

Line arrays are often used for PA systems, and you can find them in halls, theatres, cinemas, etc. You can make them more directional by building in time delays to some of the speakers (as happens insome sound bars). Unless you are using them in a field, Ideally they should be designed for for the shape and size of your room. If set up correctly they do work, by adding constuctively to the direct sound, while the timing of the echos is effectively spread, smearing them out. All you need is money for a specially designed set. I'm not so sure about off-the-peg ones.

Andrew

 

[oldric_naubhoff]

(internetless holiday second half of August -- my therapist says it's good for me!).

I guess I need to have such a therapy myself (now where would I go? maybe central Algeria? :? )

I've now done some reading, mostly on the JBLs (I'm somewhat put off by the DIY aspect of the CBT36). Apart from the CBT100LA, there's also a CBT50LA and a CBT200LA, with half and double amount of drivers respectively. They have the same freq. response, the most obvious difference being lower/higher SPLs, but I notice some more differences. The vertical coverage is fixed at 20° in the 50 (I guess that won't be too much of a problem). But then they mention the 'Pattern Control Frequency', at 1500 and 600 Hz for the 50 and the 100 respectively. Does that mean the Constant Bandwith Technology only works properly from this freq. and up?

correct. but it's 1500Hz and 1000Hz. that's why 100s should be a better bet. like I mentioned in your other thread; I need to confirm with JBL their CBTs can be used in apartments, rather than only in large spaces. I know using a CBT array in apartment shouldn't be a problem with CBT36 because I saw measurements taken a few inches in front of the baffle and then a few metres away and they virtually don't change save for a slight decrease in overall level, but that's quite understandable that volume decreases with distance. however, with array this is much less than in case of a typical point source or quasi point source (like 2-way bookshelf). typically you loose 6dB of output for every doubling of distance with a point source and only 3dB with a line array. and this is a good news for your amp if you believe in superiority of power quality over quantity. I'm just hoping that in case of JBL's arrays they perform similarly nearfield which would make them suitable for home use.

the 200 looks interesting too but I'd confirm first if such a long array would work well in a room with so narrow vertical boundaries. I presume your room is typical 2.5m high. but if it could be used that way it literally means a wall of sound in your apartment! say bye-bye to shrunken stereo images.

 

[DocG]

Like I mentioned in your other thread; I need to confirm with JBL their CBTs can be used in apartments, rather than only in large spaces.

I just dropped the Belgian distributor a mail. Let's see who answers first.

I know using a CBT array in apartment shouldn't be a problem with CBT36 because I saw measurements taken a few inches in front of the baffle and then a few metres away and they virtually don't change save for a slight decrease in overall level, but that's quite understandable that volume decreases with distance.

I guess it wouldn't be wise to get the CBT36 as a first DIY experiment; still toying with the idea to have it finished and put together by my carpenter and electrician. That would raise the cost, but it might still be acceptable (JA in Stereophile quotes $8500 for a finished pair!)...

 

[oldric_naubhoff]

As well as good dispersion, line arrays also give you better power handling.

truth is most line arrays don't offer all that good dispersion. especially those with evenly driven drivers. you get severe lobes off axis that don't contribute to good sound quality.

Also for nearfield listening you get comb effects.

the beauty of CBT is that you don't really. you essentially get the same freq response regardless of distance from the speaker, save for slight drop in level further down the room. that means inches and metres away from the baffle you get the same sound.

And there's nothing new about them. They've been around since at least the 1960's.

quite right but CBT is based on military research in the 70's which were declassified in the 80's. it took another few decades for some forward thinking people to come up with an idea that this research could be used for building speakers. IIRC Don Keele started to work on CBT prototype around 2000.

 

[oldric_naubhoff]

I guess it wouldn't be wise to get the CBT36 as a first DIY experiment; still toying with the idea to have it finished and put together by my carpenter and electrician. That would raise the cost, but it might still be acceptable (JA in Stereophile quotes $8500 for a finished pair!)...

AFAIK finished speakers are not available yet and I don't know if they ever will be since no sign of them so far.

as for assembling the kit; I guess you won't need a carpenter because the cabinets will come in fully assembled. as for electric work inside it's very easy, you just need to connect the drivers and resistors as required but the manual is said to be very thorough. it's just the bulk of work that may seem overwhelming. the only tough part where you might need some help is checking if the drivers' phasing is OK for which you'll need some extra electric testing equipment and then painting of the cabinet (if you want good effect go for car paints but then you'll defo need professional help) or veneering (in which case a carpenter might be useful).

if you're interested you might check out this thread where a guy reports how he dealt with the kit:

http://www.amb.org/forum/new-speaker-fi-audio-artistry-cbt36-t1887.html?sid=38c7a6593838355c8ccd20b006df2f89

 

[sheggs]

I presume the manufacturers haven't heard of room modes (or physics)
Even if the speakers self eq'ed how would they do this?
I wonder if they make you more attractive also?

 

[lindsayt]

Also for nearfield listening you get comb effects.

the beauty of CBT is that you don't really. you essentially get the same freq response regardless of distance from the speaker, save for slight drop in level further down the room. that means inches and metres away from the baffle you get the same sound.

I'm wondering by what sort of witchcraft or sorcery these CBT speakers avoid nearfield comb effects?

Did they not do the schoolboy physics experiment where you put 2 vibrating paddles in a dish of water?