Jason Allen looks at loudspeakers and power amplifiers, which are respectively the most customer-facing and most hidden elements of your installation.

The selection, installation and optimisation of loudspeakers will have the single biggest impact on the quality of any job.

The perceived performance of the entire system will largely be a product of how well the loudspeakers reproduce signal, cover the required area and at what volume. With a bewildering number of options available for any size of job, the most important thing to know when deciding on which way to go is what your customer’s needs and expectations are.

There’s a huge difference in the equipment and budget required between simply reinforcing speech intelligibility and amplifying full range music at high volumes.

Even if the PA will only be used at high volumes for full bands or orchestras occasionally, it has to be designed to that specification if the customer expects it to perform.

After your customer’s expectations, the next most important factor is the nature of the space itself – its dimensions, the audience area, what it’s made of, how noise affects its neighbours, how reverberant it is and any restrictions on weights, access and cabling. These factors may or may not conflict with your customer’s expectations. For example, it’s going to be very difficult to provide high levels of speech intelligibility or coherent amplification of a rock band in a highly reverberant space such as a traditional stone Christian cathedral.

MEETING THE SPEC
Depending on the customer and if they’ve received any professional advice from engineering consultants, they may provide you with a list of performance criteria that any PA system installed must satisfy.

These can come in a few forms. One is a volume requirement, for example ‘the system must be capable of delivering an average of 98dB SPL to each seat’. Another might be STI, or ‘Speech Transmission Index’, a method of measuring a PA’s  level of speech intelligibility by assigning a value between 0 and 1, with 1 being excellent. An STI specification will usually read ‘the system must have an average STI of 0.55’ or similar. The onus is then on you to select equipment that can meet the desired performance, optimise it to do so and then measure and document the fact that it does.

There are a variety of predictive software products available that can help you model the acoustic response of a space given the installation of specific loudspeakers.

The industry leader is EASE (http://ease.afmg.eu/) though this is an expensive and complicated program for the casual user. Its advantages are that it provides detailed, realistic simulations, its predictions are widely accepted by architects and consultants, and data on almost every reputable loudspeaker is available for use within the program. Most  established loudspeaker manufacturers also have their own software, usually free, that predicts room response. The obvious limitation is that they only include data on their own models.

Many distributors and wholesalers are happy to assist with software modelling for a system and are best placed to understand the response of their own products in any given installation. They are an invaluable source of advice for
any install and will have several useful suggestions for products that fit your use case.

LINE SOURCE VS POINT SOURCE
Until the late 1990s, most loudspeakers for speech and music amplification in performance spaces were what is known as ‘point source’. This is the loudspeaker design most of us are familiar with – a single cabinet with a woofer and a compression driver, with the woofer either 10″, 12″ or 15″ and the compression driver behind a horn or waveguide with 1″ exit diameter that shapes the mid-high frequencies to propagate at a set angle. These designs are called ‘point source’ because the sound emanates from their single point in space and spreads pretty much evenly from there, declining in volume according to the inverse square law.

Any given point source loudspeaker will cover the set area you point it at in a predictable range of frequencies and volumes, spreading evenly out in a spherical shape, declining in volume relatively quickly the further away you get. Want more coverage? Add more boxes. The trick with point source loudspeakers is ensuring you have enough in the right places, not interfering with each other. The points at which their coverages cross over are inevitable and cause problems.

Time alignment and EQ are the keys to optimising a point source system. Line source, or line array, while not a new concept, was revived for PA applications in the mid-1990s by L’Acoustics and has grown exponentially in popularity since. A line array is a long vertical column of many drivers acting together. The benefit of using many drivers arrayed vertically is directivity – sound propagates out from a line array in straight line, not the spherical wavefront of point source. This means it is easier to steer the sound exactly where you want it to go, instead of having acoustic energy reflect back off of ceilings and floors. It also means the volume drop-off at distance is around half that of point source.

The frequency range in which directivity is controlled is a physical consequence of the array’s length. The lowest frequency that the array can control the directivity of is L = A/2, where A = wavelength and L = array length.

As a guide, the lowest frequency we can hear is 20Hz, which has a wavelength of 17m! Most line arrays typically have pattern good control from around 100Hz up, but there is also a high frequency limit on pattern control. At higher frequencies, the limiting factor is how close together the individual drivers are, which to achieve pattern control is around half the wavelength. Considering that 6kHz is 57mm long, drivers would need to just 28.5mm apart. This is all important to keep in mind when being sold a line source system. There are many installs out there with ‘line arrays’ that are a mere one metre or so long. Not really a line array in the functional sense of the word, at least not a particularly wide range of frequencies.

So which is better?

Choosing between line and point source comes down to the physical nature of the room and budget. Line sources are long and need to be hung from the ceiling, in most cases, to cover the audience properly. As they use many drivers and cabinets, they also need a lot of amplification and DSP, which means greater expense. Point source, by contrast, is useful for covering spaces with many nooks and crannies, such as traditional stalls/circle/balcony theatres or houses of worship. They can be easier to rig and use less channels to amplify, making them cheaper, but ultimately are more difficult to tune and configure as evenly as a line array.

ACTIVE VS PASSIVE
Another consideration in loudspeakers is whether to use active (the amplifier is built into the cabinet) or passive (needs a separate power amplifier) models. This decision applies to both point and line source systems. The advantage of active is that you only need to supply power and signal to the box, which is usually easy and cheap to access or patch.

The downside is that they are heavier, more expensive, and harder to replace or repair in the case of an amplifier fault.

Passive loudspeakers are lighter and cheaper, but you need to buy separate power amplifiers and processing to run them. You also need to take into account speaker cable, which is heavy gauge copper and needs to be run from the power amplifiers to the loudspeakers. It can be expensive, and there are limitations on how far you can run it before you start to lose too much power.

Passive loudspeakers offer great flexibility with how they are powered and processed – you can use any brand of amplifier or processor you like, and can often choose to run them in bi or tri amplified mode, giving even greater control. Active speakers lock you into the built-in amplifier and usually run their own processing, but you can still put in extra DSP upstream. In the case of amplifier failure, a passive system lets you simply swap out one amp in a rack. An active fault means getting to, removing and replacing an entire speaker. On the plus side with active, you don’t need to calculate which amplifier wattage you need, or carefully check how many loudspeakers you can run off of one amp channel as the decision has been made for you.

So which is better?

Again, it’s going to come down to the physical requirements of the room and the budget. It might be a cinch to install a rack of power amplifiers into a void space, get signal to it and run speaker cables to a passive system in a performing arts type venue with a stage and wings. Or the same task might be a nightmare in a reception centre with no space for racks, no way to run huge amounts of thick speaker cable, but with an abundance of power sources.

THE BOTTOM LINE
If you’re amplifying almost any kind of music, you’re probably going to need to reproduce frequencies under 100Hz, possibly at high SPL. This is where the subwoofer comes in. Loudspeakers try to be faithful things – if you ask them to reproduce a frequency, they do their level best. The problem is it’s pretty much physically impossible for just one cabinet and a set of drivers to accurately reproduce frequencies across the full range of human hearing.

Low frequencies in particular take a lot of movement and energy to project. Subwoofers are physically large out of necessity and take a lot of power amplifier wattage to run. Installing them on the floor and abutting to walls and other surfaces increases their efficiency due to a phenomenon known as loading. Their location is not as important as mid-high boxes, as we don’t perceive very low frequencies as directional, but their relationship in space to the mid-high boxes is, as they will need to be time aligned with the rest of the system to form a coherent wavefront.

OTHER SYSTEMS – HIGH IMPEDANCE
The less glamorous cousin of FOH PA, the high impedance (or more correctly, constant voltage system) is the mainstay of supermarkets and function centres everywhere. Also known as 100V (or 70V) line systems, most ceiling speakers run in this fashion. Basically, high impedance systems are a way of running many loudspeakers off of one amplifier at low cost, with minimal cabling and with a simple calculation to determine the capacity of each amp channel. The trade-off is sound quality, but for speech and background music-only systems, this is usually a compromise worth making.

High impedance amplifiers either run extremely high voltage at their outputs natively or have a step-up transformer in-line at the output stage. The higher voltage (typically 100V in Australian systems) means that more power can be carried on a thinner cable over longer distances.

High impedance loudspeakers then connect via a step-down transformer to take just as much power as they need. Most modern high impedance loudspeakers let you choose from multiple power settings. Having these transformers in-line means the maths of calculating the load on the amplifier are much simpler than in low impedance systems. If you have a high impedance amplifier that can deliver 500W per channel, you can connect 10x 50W or 20 x 25W high impedance loudspeakers (though it’s a good idea to leave 10% or 20% margin for load and peaks).

The other advantage is that you daisy chain all the loudspeakers together along one line of relatively thin and cheap two-core speaker cable. The most common examples of 100V line products are typically ceiling loudspeakers and surface mount loudspeakers. High impedance usually implies fairly low wattage and low level, and as such is sued for putting a bit of sound in a lot of places. Driver size usually ranges between 3” and 8” at the upper end, depending on models and power.

100V line systems are often used in the same installation as high powered low impedance PA systems. A church or theatre might need a serious line array in the auditorium, but will also need ceiling loudspeakers in the foyers for audience paging.

Some manufacturers have responded to this market demand by producing power amplifiers that can switch from high to low impedance as desired, making stocking spares and replacing damaged units simple, something well worth keeping in mind when you have to service both types of system.