Meeting the Sonductor
Steve Harris talks to Danish engineer Ole Lund Christensen about his techniques for improving the performance of the listening room in the sound of a serious Hifi system
At a show many years ago I met a hi-fi enthusiast who was upset and puzzled because his expensive system no longer sounded as good as it had before, and in fact now sounded awful. He’d previously been perfectly happy with it, and hadn’t changed a thing. What could have gone wrong? It turned out that he’d just got married, moving out from his parents’ solid, high-ceilinged Victorian villa, and into a modern estate house. Only then, when his system failed to satisfy the way it had before, was he forced to realise what we now all know – that the room itself makes a considerable contribution to the sound that we hear.
Every room is different. Some seem able to make almost any hi-fi system enjoyable, just as they provide a pleasant acoustic environment for relaxation and conversation when music is not being played. Others have the opposite effect, never seem to be comfortable, and in them the speakers never seem to ‘sing’. Of course, any room can be improved, at least to some extent, with acoustic treatment of various kinds. But the ultimate solution – only a dream for most of us – is a purpose-built listening room with ideal acoustics.
One engineer who has shown how this dream can be made a reality is Ole Lund Christensen from Denmark. Christensen has a background in recording studio design and professional audio, but in the last few years he has worked extensively with domestic systems and rooms. Recently, I’ve been able to follow the progress of the room Christensen has built for an enlightened UK client, who gave him a virtual carte blanche to create the best possible listening environment within a given space.
At the outset, the brief from a client who had just moved house was to create a room that would maximise the potential of his existing high-end twochannel stereo music system. It also had to provide a multi-channel environment for watching movies. The first requirement, says Christensen, is silence, as small details will simply not be audible without a silent background. “Recording studios, and the BBC, have known this for years, and they have standards. The biggest problem is actually the air conditioning!” For this client’s installation Christensen created a room-within-a room, with a floating floor and up to a foot of rockwool between the structural walls and the cloth ‘walls’ of the listening room itself. And the air conditioning has been carefully designed with lengthy ducting so that it is truly inaudible in the listening space. With soundproofing almost to recording-studio standards, no sound from the room is heard in the rest of the house. However, it wasn’t possible to prevent some bass leakage to the room below within the limited height available and when the system is being played very loudly. But as the client says: “When I’m in here, I can’t be disturbed!”
It seems obvious that when we listen to music at home, we want to hear voices and instruments exactly as they were recorded. So one would think we need to avoid any acoustic contribution from the listening room. Don’t we just want to hear the sound coming directly from the speakers, and absolutely nothing else? Well, almost, but not quite…“Anechoic rooms have been around for a long time as a research tool, and people have naturally put up a pair of speakers and sat there in a chair,” says Christensen. “And it sounds dreadful! It’s a very artificial environment. So it’s totally counter-productive to the enjoyment of music.”
In reality, we need a room whose acoustics feel comfortable when we walk in and sit down, but at the same time one which allows the loudspeakers to present both the recorded sound and its acoustic unimpeded. For analysis, the sound from a loudspeaker in a room can be divided into three parts. First, there is the direct sound from the speakers. Next come the early reflections, sounds that in a small room will arrive at the listener’s ear just a few milliseconds later, having bounced off the side walls, the floor or the ceiling. Finally, the sounds arrive after reflecting around the room for rather longer before reaching the listeners ear. In a large concert hall they will have reflected long enough to become diffuse and form the reverberation; long first reflections provide the information that one is in a big space.
In an average home listening room, it’s likely that more than half of the sound energy that reaches the ear is coming as reflections from the room surfaces, not directly from the loudspeaker. The brain is attempting to focus on the direct sound, but is continually having to deal with the reflected sound. “Actually these ‘normal’ listening rooms that we have are a nightmare for the brain”, says Christiansen, “And only because we have such a huge amount of processing power, and it works so well, are we not conscious of it.
“We all know about the ‘cocktail party effect’. You can have a very noisy room full of chattering people, but if somebody mentions your name, aah! You hear it! So our brain has the ability to pick out tiny information from a very cluttered environment. But it’s hard work. So I worked on reducing these reflections as much as possible, only allowing a limited amount of final diffuse reverb.”
In a big concert hall, the distances from the sound source to the wall and then back to the listener are very large, but in a small room it’s different story. “When the sound goes out it gets reduced in level by the square for every doubling of the distance. So the reflections in the concert hall, because it’s so big, and the distances are so big, are very weak. But the thing that you hear in a small room is the fact that the reflections are very loud. So I said, get rid of them! Absorb them!”
Essentially, Christensen has produced a listening space that feels and sounds much bigger than it is. The client’s room measures around 6 x 4m, with the speakers facing the listener along the shorter axis. At the sides, the corners are cut off by facets at carefully-chosen angles, the rear ones concealing the rear speakers. Directly at the sides, between these corner sections, there are reflective surfaces formed by the substantial (2000kg!) military brick housings for the subwoofers, concealed only by the cloth covering. But these too are carefully angled, actually diverging by 6 degrees, and in fact there are no parallel surfaces in the room. All the other surfaces except the floor are absorbent, the back wall especially providing what Christensen calls a ‘black hole’ to deal with early reflections.
“If you look at the four 15 inch subwoofers, you’ll see a very nice hard brick surface. But, if you look at where the [front] speakers are, they reflect into the absorber. So I make sure all my first reflections are absorbed. They are not coming towards you. The sound has to go at a rather awkward angle to reach here, it must bounce off that, and even then it won’t hit you, but might arrive over there. It has to undergo many, many hard surface reflections before it gets to you.
I am actually making sure that the reflections are travelling a long way, just like a concert hall. By the time the reflections get to the listener, they will be diffuse and at a low level. I’m trying to create an acoustic in here which is as close as I can get to a big concert hall, but with a much shorter reverb time. I’m using the laws of physics and geometry to delay and reduce those first reflections, because they are the main problem to fight. But I have to have some diffuse sound, otherwise it’s not natural.
“In a normal or even a quite dead room, with say 0.5 seconds reverberation time, and in a typical listening position of say 3m from the speakers, you actually get more power from the walls and ceiling than from the direct sound. Technically, we say then that you are beyond the critical distance. This is the distance where the power from the direct sound, and the power from all other surfaces, are equal. Most people, in most rooms, are listening far beyond that point.
“But in this room, you are inside the direct dominating field. The reverb time of this room is 0.09 seconds, which is very close to anechoic, but it is not anechoic. You still sense this tiny amount of reflection, which is intentional. This comes partly from the cloth wall covering, which is in fact very slightly reflective. Also, the thin wooden battens seen on the cloth have a scattering effect at high frequencies, helping the diffusion.
With the room essentially completed, and after various upgrades to the electronics, the client’s Audio Physic Medea speakers proved to be the weakest link in the system. Excellent though they were in many ways, the Medeas, with their Manger bending-wave drive units, just did not go loud enough. “We realised that the client was playing some of his classical music so loudly that some distortion was coming from the Mangers. This is partly due to the fact that this room is so dead, so you are basically only listening to direct sound in this room, which is unusual.”
This realisation cleared the way for Christensen to pursue his ideal approach of completely integrating the speakers into the room. So the Medeas were moved out and replaced by custom-built active main speakers, which Christensen built into the wall. “My good friend, the late Poul Ladegaard of Bruel & Kjaer, put it in a very nice way. He said, we can’t discuss the loudspeaker without discussing the room. They are interacting. Then I said, well, how do I solve this problem of the interaction?
Interaction means that the loudspeaker is basically sending sound in all directions, high frequencies directional, maybe, but lower down it gets more omnidirectional. So I thought, we can keep on trying to build loudspeakers that try to solve the room problem. But the ultimate solution is not to have the problem! With a speaker in the room, you have direct sound, and then you have sound that goes to the wall and is reflected back. So I’m solving the nearest-wall reflection problem by putting the speaker into the wall, so it can’t interact. I’m solving the other walls by absorbing.”
A key element here was the use of a rather special high-sensitivity tweeter, made by Dutch pro-audio specialist Stage Accompany and based on an ingenious Philips patent. Christensen had already created rear-channel in-wall speakers, each combining one Manger drive unit with a pair of 8 inch Scanspeak bass units, to give the best possible tonal match for the Medeas at the front. Now these rear speakers were also rebuilt with the Stage Accompany tweeters, to match the new front speakers.
Though often categorised as a ribbon, the Stage Accompany unit is actually a planar type, with conductors just like very thin circuit board tracks, printed onto a thin plastic diaphragm. “This means without losing field strength. Because you actually have two gaps.
With a radiating area that’s doubled up, the tweeter can handle more power, and have less excursion and therefore less distortion. “It’s meant for very expensive PA systems. And there, remember, such a tweeter will have to send sound into a very big room, entertaining thousands of people. So you really need a lot of power. This tweeter can actually create a louder undistorted signal than any compression driver, simply because the area is so big.
The distortion of a horn loudspeaker (a compression driver), is actually not coming from the magnet or the voice-coil, it’s that the air itself, being compressed so much with so much power, becomes non-linear. Very simply, you have one atmosphere of pressure. You can go up to hundreds of atmospheres, but you can only go negative to zero. It’s inherently non-linear. You can compress, but you can’t rarefy enough. So you get one-sided distortion, which is the second harmonic. Correctly implemented, these tweeters can become almost perfect, and are certainly far better than any normal ribbon tweeter.
At the extreme bass end, Christensen’s angled walls increase the radiation impedance seen by the woofer, and so help to reduce distortion: “Normally, when a woofer is producing 25Hz, it will have to move further than it does at 50Hz. But here the radiation impedance increases as the frequency goes down, so it doesn’t have to work so hard.”
After all this, I can only say that listening to the client’s system is a wonderful experience. Behind those in-wall speakers, the electronics now include Audiocom’s Oppo Signature player and a 16-channel Trinnov processor, providing time-aligned digital crossovers at 1.2kHz for the five tweeters, all units being driven by Christensen’s own custom-built power amplifiers.
On the latest Blu-ray music discs, it sounds simply magnificent, as the music becomes totally absorbing and involving. I listened all day and stayed overnight! What’s almost equally striking is the way that this system can give an organic, joyful listening experience with ordinary CD-quality material, or even (believe it or not) the occasional MP3 track. Christensen’s efforts do seem to have swept away the usual constraints of a domestic room, bringing much of the easy, immediate and natural quality of live music into a home environment.
And even if you aren’t about to undertake a major room-building project, there’s a lot to be learned from his approach. In the end, as he puts it, the speaker and the room simply have to work together. “Like a man and a woman, you could say… No, let’s just say, they must work together in a harmonious partnership.”
Source: HIFI Critic / Audio Review Magazine, vol.9/No2 Jul – Sept 2015
At a show many years ago I met a hi-fi enthusiast who was upset and puzzled because his expensive system no longer sounded as good as it had before, and in fact now sounded awful. He’d previously been perfectly happy with it, and hadn’t changed a thing. What could have gone wrong? It turned out that he’d just got married, moving out from his parents’ solid, high-ceilinged Victorian villa, and into a modern estate house. Only then, when his system failed to satisfy the way it had before, was he forced to realise what we now all know – that the room itself makes a considerable contribution to the sound that we hear.
Every room is different. Some seem able to make almost any hi-fi system enjoyable, just as they provide a pleasant acoustic environment for relaxation and conversation when music is not being played. Others have the opposite effect, never seem to be comfortable, and in them the speakers never seem to ‘sing’. Of course, any room can be improved, at least to some extent, with acoustic treatment of various kinds. But the ultimate solution – only a dream for most of us – is a purpose-built listening room with ideal acoustics.
One engineer who has shown how this dream can be made a reality is Ole Lund Christensen from Denmark. Christensen has a background in recording studio design and professional audio, but in the last few years he has worked extensively with domestic systems and rooms. Recently, I’ve been able to follow the progress of the room Christensen has built for an enlightened UK client, who gave him a virtual carte blanche to create the best possible listening environment within a given space.
At the outset, the brief from a client who had just moved house was to create a room that would maximise the potential of his existing high-end twochannel stereo music system. It also had to provide a multi-channel environment for watching movies. The first requirement, says Christensen, is silence, as small details will simply not be audible without a silent background. “Recording studios, and the BBC, have known this for years, and they have standards. The biggest problem is actually the air conditioning!” For this client’s installation Christensen created a room-within-a room, with a floating floor and up to a foot of rockwool between the structural walls and the cloth ‘walls’ of the listening room itself. And the air conditioning has been carefully designed with lengthy ducting so that it is truly inaudible in the listening space. With soundproofing almost to recording-studio standards, no sound from the room is heard in the rest of the house. However, it wasn’t possible to prevent some bass leakage to the room below within the limited height available and when the system is being played very loudly. But as the client says: “When I’m in here, I can’t be disturbed!”
It seems obvious that when we listen to music at home, we want to hear voices and instruments exactly as they were recorded. So one would think we need to avoid any acoustic contribution from the listening room. Don’t we just want to hear the sound coming directly from the speakers, and absolutely nothing else? Well, almost, but not quite…“Anechoic rooms have been around for a long time as a research tool, and people have naturally put up a pair of speakers and sat there in a chair,” says Christensen. “And it sounds dreadful! It’s a very artificial environment. So it’s totally counter-productive to the enjoyment of music.”
In reality, we need a room whose acoustics feel comfortable when we walk in and sit down, but at the same time one which allows the loudspeakers to present both the recorded sound and its acoustic unimpeded. For analysis, the sound from a loudspeaker in a room can be divided into three parts. First, there is the direct sound from the speakers. Next come the early reflections, sounds that in a small room will arrive at the listener’s ear just a few milliseconds later, having bounced off the side walls, the floor or the ceiling. Finally, the sounds arrive after reflecting around the room for rather longer before reaching the listeners ear. In a large concert hall they will have reflected long enough to become diffuse and form the reverberation; long first reflections provide the information that one is in a big space.
In an average home listening room, it’s likely that more than half of the sound energy that reaches the ear is coming as reflections from the room surfaces, not directly from the loudspeaker. The brain is attempting to focus on the direct sound, but is continually having to deal with the reflected sound. “Actually these ‘normal’ listening rooms that we have are a nightmare for the brain”, says Christiansen, “And only because we have such a huge amount of processing power, and it works so well, are we not conscious of it.
“We all know about the ‘cocktail party effect’. You can have a very noisy room full of chattering people, but if somebody mentions your name, aah! You hear it! So our brain has the ability to pick out tiny information from a very cluttered environment. But it’s hard work. So I worked on reducing these reflections as much as possible, only allowing a limited amount of final diffuse reverb.”
In a big concert hall, the distances from the sound source to the wall and then back to the listener are very large, but in a small room it’s different story. “When the sound goes out it gets reduced in level by the square for every doubling of the distance. So the reflections in the concert hall, because it’s so big, and the distances are so big, are very weak. But the thing that you hear in a small room is the fact that the reflections are very loud. So I said, get rid of them! Absorb them!”
Essentially, Christensen has produced a listening space that feels and sounds much bigger than it is. The client’s room measures around 6 x 4m, with the speakers facing the listener along the shorter axis. At the sides, the corners are cut off by facets at carefully-chosen angles, the rear ones concealing the rear speakers. Directly at the sides, between these corner sections, there are reflective surfaces formed by the substantial (2000kg!) military brick housings for the subwoofers, concealed only by the cloth covering. But these too are carefully angled, actually diverging by 6 degrees, and in fact there are no parallel surfaces in the room. All the other surfaces except the floor are absorbent, the back wall especially providing what Christensen calls a ‘black hole’ to deal with early reflections.
“If you look at the four 15 inch subwoofers, you’ll see a very nice hard brick surface. But, if you look at where the [front] speakers are, they reflect into the absorber. So I make sure all my first reflections are absorbed. They are not coming towards you. The sound has to go at a rather awkward angle to reach here, it must bounce off that, and even then it won’t hit you, but might arrive over there. It has to undergo many, many hard surface reflections before it gets to you.
I am actually making sure that the reflections are travelling a long way, just like a concert hall. By the time the reflections get to the listener, they will be diffuse and at a low level. I’m trying to create an acoustic in here which is as close as I can get to a big concert hall, but with a much shorter reverb time. I’m using the laws of physics and geometry to delay and reduce those first reflections, because they are the main problem to fight. But I have to have some diffuse sound, otherwise it’s not natural.
“In a normal or even a quite dead room, with say 0.5 seconds reverberation time, and in a typical listening position of say 3m from the speakers, you actually get more power from the walls and ceiling than from the direct sound. Technically, we say then that you are beyond the critical distance. This is the distance where the power from the direct sound, and the power from all other surfaces, are equal. Most people, in most rooms, are listening far beyond that point.
“But in this room, you are inside the direct dominating field. The reverb time of this room is 0.09 seconds, which is very close to anechoic, but it is not anechoic. You still sense this tiny amount of reflection, which is intentional. This comes partly from the cloth wall covering, which is in fact very slightly reflective. Also, the thin wooden battens seen on the cloth have a scattering effect at high frequencies, helping the diffusion.
With the room essentially completed, and after various upgrades to the electronics, the client’s Audio Physic Medea speakers proved to be the weakest link in the system. Excellent though they were in many ways, the Medeas, with their Manger bending-wave drive units, just did not go loud enough. “We realised that the client was playing some of his classical music so loudly that some distortion was coming from the Mangers. This is partly due to the fact that this room is so dead, so you are basically only listening to direct sound in this room, which is unusual.”
This realisation cleared the way for Christensen to pursue his ideal approach of completely integrating the speakers into the room. So the Medeas were moved out and replaced by custom-built active main speakers, which Christensen built into the wall. “My good friend, the late Poul Ladegaard of Bruel & Kjaer, put it in a very nice way. He said, we can’t discuss the loudspeaker without discussing the room. They are interacting. Then I said, well, how do I solve this problem of the interaction?
Interaction means that the loudspeaker is basically sending sound in all directions, high frequencies directional, maybe, but lower down it gets more omnidirectional. So I thought, we can keep on trying to build loudspeakers that try to solve the room problem. But the ultimate solution is not to have the problem! With a speaker in the room, you have direct sound, and then you have sound that goes to the wall and is reflected back. So I’m solving the nearest-wall reflection problem by putting the speaker into the wall, so it can’t interact. I’m solving the other walls by absorbing.”
A key element here was the use of a rather special high-sensitivity tweeter, made by Dutch pro-audio specialist Stage Accompany and based on an ingenious Philips patent. Christensen had already created rear-channel in-wall speakers, each combining one Manger drive unit with a pair of 8 inch Scanspeak bass units, to give the best possible tonal match for the Medeas at the front. Now these rear speakers were also rebuilt with the Stage Accompany tweeters, to match the new front speakers.
Though often categorised as a ribbon, the Stage Accompany unit is actually a planar type, with conductors just like very thin circuit board tracks, printed onto a thin plastic diaphragm. “This means without losing field strength. Because you actually have two gaps.
With a radiating area that’s doubled up, the tweeter can handle more power, and have less excursion and therefore less distortion. “It’s meant for very expensive PA systems. And there, remember, such a tweeter will have to send sound into a very big room, entertaining thousands of people. So you really need a lot of power. This tweeter can actually create a louder undistorted signal than any compression driver, simply because the area is so big.
The distortion of a horn loudspeaker (a compression driver), is actually not coming from the magnet or the voice-coil, it’s that the air itself, being compressed so much with so much power, becomes non-linear. Very simply, you have one atmosphere of pressure. You can go up to hundreds of atmospheres, but you can only go negative to zero. It’s inherently non-linear. You can compress, but you can’t rarefy enough. So you get one-sided distortion, which is the second harmonic. Correctly implemented, these tweeters can become almost perfect, and are certainly far better than any normal ribbon tweeter.
At the extreme bass end, Christensen’s angled walls increase the radiation impedance seen by the woofer, and so help to reduce distortion: “Normally, when a woofer is producing 25Hz, it will have to move further than it does at 50Hz. But here the radiation impedance increases as the frequency goes down, so it doesn’t have to work so hard.”
After all this, I can only say that listening to the client’s system is a wonderful experience. Behind those in-wall speakers, the electronics now include Audiocom’s Oppo Signature player and a 16-channel Trinnov processor, providing time-aligned digital crossovers at 1.2kHz for the five tweeters, all units being driven by Christensen’s own custom-built power amplifiers.
On the latest Blu-ray music discs, it sounds simply magnificent, as the music becomes totally absorbing and involving. I listened all day and stayed overnight! What’s almost equally striking is the way that this system can give an organic, joyful listening experience with ordinary CD-quality material, or even (believe it or not) the occasional MP3 track. Christensen’s efforts do seem to have swept away the usual constraints of a domestic room, bringing much of the easy, immediate and natural quality of live music into a home environment.
And even if you aren’t about to undertake a major room-building project, there’s a lot to be learned from his approach. In the end, as he puts it, the speaker and the room simply have to work together. “Like a man and a woman, you could say… No, let’s just say, they must work together in a harmonious partnership.”
Source: HIFI Critic / Audio Review Magazine, vol.9/No2 Jul – Sept 2015