Currently many conference sound reinforcement systems use ceiling speaker design, especially office space round-table type of conference room. In this application using the ceiling design is obviously on the sound field of the uniformity of the index contribution is the largest. However, many projects designed by the ceiling speaker system or along the way the background music sound reinforcement connection and control, so that not only did not play the strong points of the overhead sound reinforcement system, but instead give a person a kind of "cheap project" feeling. Later in this article, we take a meeting room of the State Council as an example of the actual project, to introduce the reader to the top of the sound reinforcement system of all the advantages.

We first want to make it clear that the conference sound reinforcement in the end how to be considered good? I think we should consider from two aspects, one is to talk about the function, that is, the advanced nature and flexibility of the issue, but also simple operation, this issue is not the focus of the discussion in this paper; the second is the effect of the sound field, for the conference sound reinforcement system we care about what? The answer is that the first to hear, followed by listening to the clear. Then it corresponds to the technical problem is to have enough sound pressure level and enough clarity, we are here to discuss the two issues around.

First, the sound field of the conference room sound and image problems

Domestic conference room pattern is roughly divided into podium-style "report conference hall" and round-table conference halls of two categories. The former is usually a large indoor space, divided into two parts of the podium and the audience. At present, many of the engineering design is the use of front main speakers, and in the side walls of the practice of hanging filler sound box, such as Figure 1:



Figure I report on the design of the main speakers in the conference hall + multiple complementary sound box

This speaker design seems to be "universal", because it does not take into account the room aspect ratio, does not take into account the height of the room, does not take into account the vertical depth, does not take into account the conference room there is no columns and other obstructions, etc., can be used. For occasions where the longitudinal depth is too large, it is only necessary to increase the delay in the rear field. The biggest problem with this design is that the uniformity of the sound field is very poor, that is, the larger the room uniformity is worse, the listener close to the speaker sound pressure is too large. And because of the uneven speaker layout will greatly limit the sound transmission gain, the same layout for the positioning of the sound image is not very accurate.

Let's talk seriously about sound positioning. At the beginning we have confirmed that the two most important issues of conference system is the sound pressure level and clarity of the problem, and did not mention the problem of the sound image. This point is completely different from the performance system, for the performance system, the sound positioning is even more important than the clarity, so almost all of the large-scale performance system, as long as there is a way through the main sound reinforcement system can be uniform coverage of the whole field design, it is absolutely not allowed to use the fill speakers, because any side of the fill sound, top fill sound will be on the sound image of the correct original impact. But in the conference system, our main task of sound reinforcement is to allow listeners to hear the speaker's speech content most clearly, rather than caring about the direction from which the sound is sent over (in the United Nations General Assembly, the audience is wearing headphones for meetings, they will not think that the speaker is to come together in his ear to speak), which means that you can hear it clearly, on the purpose of the purpose. So we in the conference sound reinforcement design, we should put the sound and image of the problem at the end.

For some of the transformation of the system, the temporary construction of the venue or do not have a ceiling mounted speakers on the occasion, I also recommend that we try to use the speaker layout shown in Figure II.



Figure 2 Fill-in speaker placement

Because in Figure 1 in the fill 3 and 4 of the two speakers is completely redundant, even if the depth of the conference room is very narrow, it is only necessary to add a pair of speakers and the same in accordance with the rear of the placement of the method of panning to the back of the field. Like Figure 1, the complementary sound speakers 3 and 4 placed in the back of the two corners, and then to the direction of the podium is very undesirable, at this point in the middle of the seat of the listener will get some delay different directions of the sound, but reduce the clarity of the sound. To know, we are not in the appreciation of multi-channel film surround sound (film surround sound in different speakers issued by the different contents of the sound, so it can produce a feeling of enclosure, but all the channels in the conference system are exactly the same sound, they can only produce interference in the space convergence). Of course, according to the placement of Figure 2 should be appropriate to increase the delay on the rear field speakers.

Another round-table type conference room is even less recommended to use the four corners of the encircling structure, such as Figure III.



Figure three round-table room of the four corners of the speaker layout will produce serious whistling

This is almost difficult to achieve sufficient sound transmission gain, the phenomenon of whistling will be very serious, and even the use of a large number of sound-absorbing materials and feedback suppressor also does not help.

Second, ceiling loudspeaker network layout programme Note

In most of the conference room sound reinforcement system, we recommend the use of ceiling loudspeaker design, due to the sound field is easy to distribute evenly, but also greatly reduces the probability of acoustic feedback, improve the sound transmission gain. Of course, in such a design, but also to give full consideration to the directivity of the conference microphone selection, usually a super-cardioid pointing gooseneck microphone is appropriate, and try to avoid omnidirectional or hemispherical interface microphone (PZM), but also try to avoid the use of highly sensitive lavalier microphone.

Ceiling speakers can have the following three arrangements according to the distribution of radiation area. See Figure IV.



Figure 4 Distribution of ceiling speakers

Used as a conference system, we only recommend the second one, which is the minimum overlap. The circle shown above is a sound field distribution map measured at the height of the listener's ear, and the corresponding number of ceiling loudspeakers installed is of course also related to the height of the room's ceiling as well as the effective radiation angle of the loudspeaker (pointing angle). In Figure 5, all we need to know to accurately calculate the radius of radiation is the angle of radiation and the height of radiation.



Figure 5 Radiation range of a ceiling speaker

This is a simple mathematical calculation, we follow this calculation method to estimate the number of speakers in a room of 100 square metres, see Figure VI

Fig. 6 Estimated number of speakers in the room

From the above chart we can find that if the use of large radiation angle of the speaker can save the number of installations, from this point of view analysis is right. However, the directivity of the speakers with large radiation angle is poor, the corresponding critical distance is small, so it is unfavourable to improve the clarity of speech, isn't it a contradiction again? We leave this topic in the fourth part of the discussion.

Ceiling speakers in the ceiling layout should basically follow the principle of symmetrical placement, unless your conference room is irregular, generally speaking symmetrical distribution is conducive to the uniformity of the sound field. Refer to Fig. 7 and Fig. 8 for two types of conference room layout methods.



Figure VII podium conference room ceiling speaker layout



Figure VIII Ceiling loudspeaker layout for round table room

In Figure 7, the return speaker of the podium is also replaced by the ceiling speaker of the ceiling at the same time, which is also very good for aesthetics.

Incidentally, to promote the speaker amplifier, it is best to use two channels to promote the cross-symmetrical speakers, such as Figure 8, with "amplifier 1" to promote the ceiling 1, 5; with "amplifier 2" to promote the ceiling 2, 6; with "amplifier 3" to promote the ceiling speakers. Amplifier 3" to promote the ceiling 3, 4, the purpose of doing so is in case of a failure of an amplifier, but also not the whole half of the sound.

Third, effectively avoid flutter echo

We know that in the closed space composed of reverberant sound field inside, the sound will resonate indoors, the resonant frequency wave is called a simple positive wave, when a simple positive wave in more than one direction at the same time overlap (acoustically known as the phenomenon of inter-parallel) will occur a strong frequency resonance, this resonance frequency is very easy to cause acoustic whistling. As the root cause of this problem is due to the physical distribution of space and the characteristics of sound wave transmission caused by the sound reinforcement system has nothing to do, so we have to acoustic treatment of the room, that is, usually said to increase the sound-absorbing materials, change the structure of the reflection and so on. The conference room sound reinforcement we are discussing here is generally a very regular room structure and decoration, which is very unfavourable to the processing of acoustic properties. A lot of the later commissioning would be less strenuous if we could suggest acoustic finishes before the renovation. However, most of the projects are not allowed to intervene until the interior decoration plan has been decided or even already finished, so it is very difficult for us to control the acoustic environment of the room.

When we arrange the loudspeakers, we should try to avoid is the flutter echo, the so-called flutter echo is also a kind of simple positive wave, that is, the sound in the two hard parallel reflective surfaces between the reflection back and forth, but the energy is attenuated very little, see Figure IX.



Figure 9: Flutter Echo

Vibrato echo can be formed between the ceiling and the ground, but also in the formation of two parallel side walls, and this standing wave is very easy to stimulate the self-excitation of the electro-acoustic system to form a positive feedback, so we have to try to avoid the vibrato echo in the design of the system, when installing the ceiling speakers, the speakers of the axial preferably avoid the hard level of the desktop, you can fall in the top of people's heads or other blank areas. In addition the floor should be laid as thick as possible carpet, the seat should also try to choose soft cloth covered soft chairs. For the ceiling if possible, but also to take care of the aesthetics at the same time to do further acoustic design (make full use of the ceiling cavity sound-absorbing structure is the best solution to low-frequency resonance, but to ask professional acoustic designers to do the design guidance).

Fourth, the speaker's bandwidth, power and directional angle on the impact of clarity

In the conference sound reinforcement system, we do not need to pursue like music playback bandwidth. Generally speaking, we can control the electronic part of the filter bandwidth will be limited to 100Hz - 8kHz is already enough. Some people may want to suggest that the conference room may also want to put music and other sources, we should not forget that the sound reinforcement system is for the meeting, music is only additional, secondary things (not to mention the 8kHz bandwidth for general music playback is enough), we can not be for the secondary things to risk the risk of whistling, because the wider the bandwidth of the electro-acoustic system is opened up, the greater the probability of whistling occurs The wider the open bandwidth of the electroacoustic system, the greater the probability of whistling. In practice, sometimes in order to achieve sufficient gain, we even compress the bandwidth to 150Hz - 4.5kHz can still get enough voice clarity. For speakers, we do not recommend the use of fixed-pressure ceiling speakers (in the next part we also mentioned that in order to further increase the system gain, we must use multi-channel partition transmission signal, which requires more than one amplifier to promote all the ceiling speakers, there is no need to use fixed-pressure speakers in this application); the second reason is that the conference system is not the use of speakers used in the background music The second reason is that the conference system is not the use of speakers used in the background music of low-power speakers, and the use of more than 100W full-range speakers or speakers (can be a combination of forms, but the optimal coaxial full-band ceiling speakers), for this high-power speakers if you want to be connected to a higher-capacity transformer will further increase the cost, but can not receive any benefits; Third, is that ordinary audio transformers will be degradation of the frequency response characteristics of the speaker, but the installation of expensive audio Transformer may be the price of the transformer itself has exceeded the speaker itself.

In the conference sound reinforcement system, the problem of power is more contradictory. On the one hand, the design of a large enough output power, on the other hand, but can not push up the volume (because of whistling). So much so that when the project acceptance, engineers are happy to play pop music, that sound pressure level can be comparable to Disco. But on the other hand? A push up the microphone on the stupid, the sound of a small poor, putter a move on the whistling, and then the responsibility of a drum brain pushed to build sound above, as if our engineers are sinusoidal, the sound is not good, we are also a victim. How many designers are really from their own body to find out what is wrong and solve it? Now a lot of designers in the design of conference room power, but also use the "radiation distance doubled every time, the sound pressure level attenuation of 6dB" this theory, the result is in a hundred square metres of conference room with thousands of watts of power speakers. To know this "times the distance attenuation of 6dB" theory is used in the free sound field, we are in the conference room is closed inside the reverberant sound field, the attenuation of acoustic energy in the times the distance may be less than 1dB (in a 120 square metres of a lecture theatre, the teacher as long as there is a 2-watt loudspeaker can be all the students to listen to the clearly). In a round-table conference room, if the average sound pressure level of the sound reinforcement system is 3dB higher than the background noise, the audience will be able to hear very clearly. So the conference sound reinforcement system designers need to do is not "how much power" things, but "can output how much power" things.

In the second part of our discussion on the ceiling speaker radiation angle on the impact of sound coverage. In fact, the radiation angle affects not only the coverage characteristics, but more importantly, its transmission characteristics, for a wide pointing angle of the speaker, the corresponding critical distance will be narrowed to a very small, so that the listener's ear height is easy to enter the reverberant region, which will greatly reduce the voice clarity indicators, in order to avoid these problems, it is necessary to reduce the height of the ceiling, which in reality is not acceptable. So choose the appropriate radiation angle is very critical, in most domestic projects, choose 60º or 90º pointing angle is recommended.

V. Using Mix Minuses to Improve Sound Transmission Gain

For the concept of Mix Minuses may be most sound engineers are still relatively rusty, in fact, this is just a technical means, the method of realisation can be varied. Specifically, it is in the ceiling speakers constitute a conference system, a microphone directly above the corresponding speakers, should try to attenuate the signal output by the microphone. In other words, each speaker overhead is playing the full audio signal, but for that microphone below that speaker, it should be subtracted from the total volume for the most part. The purpose of this is to minimise the amplification between the speakers and microphones in close proximity to each other in order to increase the transmission gain, and also because there is no need for the speaker himself (and his neighbours) to listen to too much of his own voice.

Therefore, in the implementation of Mix Minuses, the inverted insert level of the microphone below (and adjacent to) each speaker must be carefully adjusted. See Figure X.



Figure X. Principle of Mix Minuses

This is just a schematic diagram in which the mixing console can be a mixing console or an electronic processor. The inverted inputs are superimposed on the bus levels using an invert button similar to the one on a mixing console, which flips the phase by 180º.

Mix Minuses to have two main conditions: First, the ceiling speakers must be a multi-channel system, each output channel of the mixing console corresponds to a loudspeaker channel, we have to carry out the region on the channel microphone inverted superimposed; the second condition is the requirement of Figure 10 in the two input signals must be strictly the same phase delay, if the two signals are not synchronised to the mixer, then the inverted signal can not be accurately. Then the inverted signal can not be accurately "subtracted" from the synthesis level, but also fails to achieve the purpose of Mix Minuses.

Sixth, engineering cases

In the State Council, a premier conference room, the use of ceiling speakers installed, the meeting for the round table form. The conference room is about 18 metres long, about 10 metres wide, ceiling height of about 5.5 metres, a total of 10 independent channel speakers, each speaker's power is 150 watts. The control system is centred on a media matrix and uses Mix Minuses technology. In the final use, the system was able to keep 70 microphones switched on at the same time for long periods of time, with up to 11 people speaking at the same time without whistling. The system only incorporates a total of 15-band parametric EQ on the outputs, and no graphic EQ or feedback suppressor is used.