Optimising Vertical Loudspeaker Position in PA Systems
The height and angle of speakers has implications for the overall sound.
Almost without exception, height is an advantage in speaker placement. If speakers cannot be flown (which usually depends on what is available to fly them from at the venue) they should ALWAYS be placed on suitable stands, rather than on tables, chairs or beer crates. Unless stands can be placed on stage, this generally limits the available height to about 2m (placing the HF horn at about 2.5m), which is less than ideal. Also, some venues have very low ceilings, which is again less than ideal. Other techniques to raise speaker height (like putting a chair under each leg of a speaker stand) are dangerous, and should not be attempted under any circumstances.
The diagrams below show the effect of different speaker positions in a typical small hall. In the first diagram, the speakers (with a 60° vertical dispersion) have been placed on stands on the auditorium floor. The HF horns are 2m above the floor.
The most important effect of this is on sound levels. In free air, sound levels diminish at the rate of 6dB for every doubling of distance. If we take 1m in front of the speaker as our reference (0dB) point, we can see that direct sound from the speaker is 24dB lower at the back of the hall.
If the speakers are producing a level of 110dB (SPL) at 1m, the direct level at the back is only 86dB. Most rock & roll audiences would not be impressed with dinner-dance volume. To get 110db at the back, you would - literally - deafen the audience at the front.
Although reflections from room surfaces would reduce this effect (the reverberant field would raise the overall level at the back of the hall), there would still be a noticeable problem: too much direct sound at the front, and not enough at the back. This might mean the sound at the back was neither loud enough, nor intelligible.
Raising the speakers helps a little:
Here, the stage height (1m) has been added to the original 2m speaker height, giving the speakers an overall height of 3m. A listener at the front, in the same position as in the first diagram, is now off axis (reducing the HF level by 6dB), and their ears are now two metres from the speaker (reducing the level by another 6dB).
Because listeners at the back would be slightly off-axis, the direct level of −24dB shown here might be one or two decibels lower (although this would again be offset by the reverberant field). However, the front-to-back difference has been halved (from 24dB to 12dB): a substantial improvement. There are two disadvantages:
Overall, however, most audiences would consider the difference to be an improvement.
If the speakers are angled downwards slightly, we obtain further improvement (Stands should NEVER be tilted. To achieve modest downward tilt - up to about 5° - a relatively inexpensive tilt adaptor (which we can supply if you need some) is manufactured by K&M):
Here the speakers have been angled downwards at an angle of 4°. Now the 12dB difference in direct sound is exact, and the difference is further reduced by the reverberant field. Also, the reflections from the back wall are no longer directly on-axis, reducing their level on stage.
If we have stands with sufficient reach, increasing the height and angle yields further improvement:
Here the speakers have been raised another metre (increasing the overall height to 4m), and the downward angle increased to 6°. This places a listener at the front further from the speaker, as well as further off-axis. The on-stage reflection is also further reduced, and the front-to-back difference - discounting the reverberant field - is now only 7dB.
Flying the speakers takes us a bit further:
Here, the speakers have been flown at ceiling height (5m in this case), and given a 9° downward tilt. As a result, we have increased the distance between our two observation positions, and reduced the difference to 6dB.
Finally, different dispersion characteristics may provide even better results:
Here, the speakers have a 50° vertical dispersion, and the downward angle has been increased to 15°. The level difference over the front half of the auditorium is now only 1dB. This would scarcely be noticeable, and the overall level (including the reverberant field) would be effectively constant throughout the auditorium.
In the first four examples, ceiling reflections might degrade the sound somewhat in the back half of the hall. In the last example, all the initial ceiling reflections within the loudspeaker's (-6dB) angle of cover pass over the heads of the audience.
As with horizontal angles, the optimum vertical angle obviously depends on the size and shape of the room, as well as the response pattern of the speakers used. However, raising the speakers and angling them downwards will almost always be advantageous, and helps to get as much direct sound to the audience as possible and get the level and tone as constant as possible throughout the audience area.