In our world acoustics is usually about reducing the reverb time in a space, we spend a great deal of energy, and money on properly designed “stuff” to absorb and scatter the sound in our spaces. In the live performance of amplified, popular music any mention of the acoustics of a space is usually a discussion of the unwanted contribution the reflected sound in the venue is making. However, it’s important for us to bear in mind that some, but not all users of performance spaces, want to extend the reverb time in a space.
We all know that different types of music require different reverberances. Transient and bass-heavy music is a disaster in a reverberant space designed for orchestral music and conversely, a string section in a dry acoustic just doesn’t work- something I think every time I see a string quartet outdoors at a wedding…
In the studio we have the option of artificial reverb which offers a simple solution, though not one which would satisfy a classical engineer but live, the choices are more difficult.
Given that concert halls are expensive and often publicly funded, it is in their interests to be as flexible as possible, ideally able to offer the ideal acoustic environment for whatever business comes their way. Some purpose-built venues offer variable acoustics through passive means, one such example is the KKL Luzern in Switzerland.
Through the use of hydraulically operated panels, the internal volume of this concert hall can be increased by a third by opening up acoustic chambers around the hall. Remote controlled heavy curtains provide variable absorption in the acoustic chambers offering a reverb in the hall which can be varied from 2.0 seconds to 2.8 seconds. This type of passive variable acoustic treatment is impressive but has to be designed into the build of the hall, retrofitting would be extraordinarily difficult and I can’t begin to imagine how expensive it must be.
Active Variable Acoustics
So If we can use artificial reverb in the studio and in amplified live events, why can’t we use this to help extend reverb times for acoustic performances? Like all of these things, this is an old idea which was initially held back by inadequate technology but is now coming of age as the technical restrictions fall away.
A simple illustration would be a solution to an issue I have encountered when mixing live sound for string quartets and solo harpists in marquees. Very little amplification has been needed in these scenarios, usually just feeding some dry signal into the delays down the tent, but by applying reverb from a pre-fade send it is possible to (very crudely) put a little space around the acoustic performance of the players without putting any, or at most very little, dry signal out of the speakers. This is a basic example of the principle of active variable acoustics.
In 1955 Phillips made the first serious attempt at active acoustic enhancement with their Ambiophony system. This system used a magnetic tape wheel with a recording head and multiple playback heads to generate multi-tap delays which were then played back through loudspeakers around the space to extend the reverb in that space. A similar system was installed in Abbey Road's studio one using a series of delay drums similar to those used in the Binson Echorec. Although unconvincing by today's standards, the system was installed in La Scala in Milan.
The growth of installed systems was slow until the mid-80s and has risen markedly since the millennium as the high channel counts necessary to create a satisfactory system have become more achievable. The advent of AoIP systems has already made an impression on this area and will continue to do so.
3 Types Of Active Acoustics Systems
So the idea of using microphones, amplifiers and speakers to assist the natural reverb in a space has been around for a long time. At its purest, the idea is to increase the amount of sound which bounces back off the walls of a venue by using electronics. By using absorption we can decrease the reflected sound and by using active elements we can increase it even further than the hardest and shiniest of walls. Of course, very reverberant rooms can "ring" uncomfortably and an active room can go even further and feed back! These ringing issues all need to be addressed in the design of active acoustic systems.
There are three approaches to creating supplementary reverb in performance spaces:
- In Line
Regenerative systems are possibly the purist's approach, by adding reflections back into the room based on the room's original reflections it is possible to extend but maintain the character of the room's original ambience. This is sometimes known as "Assistance of Sound Field". By using multiple microphones and loudspeakers across many discrete channels it is possible to put more sound back into the room from the boundary than would happen naturally.
These systems respond very naturally, reacting equally to sounds made in all parts of the room just like real acoustic reverb. They have disadvantages in that they are prone to ringing as the speakers are covering the same area as their source mics and designers have to be careful to avoid colouration.
Because regenerative systems rely on the nature of the existing acoustic in the room they are less flexible than other types of systems, allowing the introduction of a reverb which is both louder and longer but never changing in character from the room it is installed in, and always following this louder is longer, longer is louder rule.
In line systems more closely mimic the familiar use of artificial reverb on dry signals over a PA system but extends it into the hall. The two principal differences between an in line system and a regenerative system are that an in line system synthesises reflections using digital reverb and an in line system uses microphones to pick up direct sound from the performance area and is essentially a one-way process. Because of this one-way character, only sounds made in the performance area will result in an enhanced reverb. The sound the audience make will not experience the enhanced reverb. While this may be a good thing it does detract from the authenticity of the experience.
Because of the one-way/synthesised nature of the reverb in an in line system there is far less risk of ringing and there is greater freedom in the character of the reverb generated, especially if the natural acoustic in the space is relatively dry. This makes in line systems particularly well suited to multi-use venues which host amplified and acoustic events.
Although this might sound on paper very like the string quartet example I gave earlier, the important difference between this type of system and that example is the number of channels involved. By playing back many channels of carefully set up reverb through many speakers around the venue it is possible to create very convincing sounding halls using this method which, while not as true to the principle of how acoustics work in halls, offer greater freedom than the longer/louder restriction of regenerative systems.
With a regenerative system the more energy you put back into the room the greater the chance of ringing. To alleviate this it is necessary to use more discrete channels so that each channel can run at a lower gain. This is a restriction in terms of cost and complexity. By using aritificial reverb in the regenerative channels it is possible to extend the reverb without putting more energy back into the room and raising the chance of feedback while also reducing the number of channels needed.
By introducing DSP processing into the regenerative loops it is possible to use complex EQ to flatten the response and further reduce the chance of ringing. By alleviating the ringing in each loop it is possible to raise the gain and therefore run fewer channels.
So far all of these examples are about adding to or extending the natural reverb by adding energy to the room. What about the other way? Is it possible to absorb sound and remove energy from a room using electronics? It's more difficult, but at the bottom end, it is possible. Cardioid subwoofers use out of phase drivers to increase directionality by cancelling bass. This has been done for years in PA installations and has crossed over into the studio by way of the Kii Three monitors. A recent example which caught my attention was the Active bass trap from PSI Audio which uses exactly the same principles found in regenerative reverb systems but inverts the polarity of the output to cancel against the incoming bass so that bass goes in but none comes out!