Initially, vocals can be expediently taken care of with a simple mixer amplifier; something similar in concept to an instrument amplifier. Indeed, on countless occasions in the past, guitar amplifiers have been adopted for this very task, with apparent success, because high impedance microphones deliver much the same output voltage as a guitar pickup. As guitar and vocals alike operate over a similar frequency range (viz: they are vocalising instruments), tone controls which work well in conjunction with a guitar will tend to suit vocals as well. However, a high impedance microphone prefers to 'see' an input impedance around 50k, and whilst higher values aren't instrinsically erroneous, or necessarily deleterious to the frequency response of the microphone, the high impedance (470k to 2M) seen by the microphone when looking up a guitar amplifier's input socket may give rise to a higher level of hum and noise than is necessary, especially in the case of valve instrument amps.

More insidious is the fixed 'equalisation' in many guitar amplifiers, particularly in the classic valve models. In this case, even if you set the tone controls to their central, nominally 'flat' position, the frequency response of the amplifier remains far from flat. This 'tweaking' is to provide a meaty guitar sound, of course, but its concealed and unpredictable nature is unhelpful in conjunction with vocals, where equalisation for a tasty sound is usually compromised by the need to control the threshold of acoustic feedback. Thus the result of borrowing guitar amplifiers for the fledgling vocal PAs of the mid-60s was to render the feedback-prone vocals inaudible in the face of the guitars' sudden augmentation in the shape of the first Marshall stacks. And as rock invariably demands loud vocals, a mixer-amplifier designed specifically for this task, exhibiting an essentially flat response and usable equalisation is required.

Here, equalisation brings home the key difference between amplifying guitars and vocals. The guitar isn't usually prone to feedback, thus tonal aberrations can be introduced or banished ad-lib, purely to create a suitable sound; the player + guitar + amplifier + speaker(s) being regarded as an instrument in their own right, and the concept of a flat response within this 'instrument' is clearly nonsensical. In addition, every component and parameter of the 'instrument' can serve to alter the sound — from the player's technique to the design of the speaker cabinet.

By comparison, for vocals, the overriding need to be audible over or amongst similar, sustained vocalising sounds (viz: guitars) calls for high sound levels which inevitably raise the spectre of acoustic feedback. So, in turn, high level vocals initially demand the discipline of an essentially flat frequency response throughout the vocals amplification system. At the same time, the human voice, unlike a guitar, can't be so readily tweaked — exotic bodies (voice boxes) and strings (vocal chords) aren't available on the NHS, thus the onus of generating a 'tasty' sound falls squarely on the sound system. The art of vocal PA then, is to balance the conflicting requirements of achieving high levels (tonal aberrations unwelcome) and a suitable vocal sound from a less than perfect instrument (tonal aberrations required), and the distinction between the instrument and the medium of communication is less clear cut.

Amplifying this argument(!) it's apparent that all too often, the vocal sound system is called upon to do four jobs: (1) Correct its own deficiencies; (2) correct the deficiencies in the vocals; (3) produce an output that's audible over 'walls' of guitar sound and (4) counteract its own tendency to produce howls of feedback. Now these factors interact as a foursome in a complex manner, and frequently conflict with each other. Here, the golden rule 'simplify and lighten' suggests that the number of requirements should be trimmed. This we can do by starting out with essentially flat microphones, amplifiers and speakers. From this reference point (which cancels the first requirement) we can go on to add tonal aberrations (i.e. "equalise") to achieve a good vocal sound — and help cut through the guitars. As a pair, these needs are frequently complementary in any case, so we're left with the basic two-sided conflict: vocal sound versus feedback threshold. But, of course, the inherent lack of colouration in the system will make howlround much less of a problem in any case. As a result, the sound level can be higher, and there's less need to 'equalise' for the sake of being audible over other instruments; thus the dilemma becomes self correcting.

With the above in mind, it's clearly much better to invest your funds in high quality microphones and speakers boasting an essentially flat response, rather than to eschew these for the sake of elaborate, yet expensive equalisation controls on your vocals mixer-amplifier. Using a 'flattish' sound system, a bass control centered around 100Hz to regulate the 'body' of the vocals, together with presence and treble controls, centered about 5kHz and 10kHz respectively and controlling the upper harmonics will be adequate in most cases. Four band equalisation, having an additional control in the low midrange to take care of the muddiness in close-miked vocals is another useful tool, as too is an integral multiband or graphic equaliser: but without an inherent lack of colouration in the system, these are really useless gimmicks.

Aside from such parameters as equalisation and input impedance, vocal mixer-amplifiers differ from instrument amplifiers in one other crucial respect — the arrangement of the inputs. Clearly, in the case of vocals, two or four inputs will often be in use simultaneously. In this case, inputs which are regarded broadly as alternatives and which differ greatly in their facilities and characteristics aren't helpful, as in an instrument amplifier. On the other hand, a potential trap here is the multichannel mixer-amplifier with parallel inputs. In this case, each channel features a pair of input sockets, having identical characteristics as regards sensitivity and impedance, together with common gain and equalisation controls. Although this arrangement seems attractive, in that the mixer-amp's fascia is filled with a myriad of input sockets, the concept is, at best, a compromise; control over each pair of microphones is limited by the shared controls, a situation that is highly unsatisfactory, except possibly in the case of tightly disciplined close harmonies. The moral here is to be sure that you buy a usable six channels rather than three pairs of channels.

Another, more amenable species of twinned input is one providing sensitivities (circa 1-10mV/30-100mV) and input impedances (600 ohms to 1k/50k) tailored to both low and high impedance microphones respectively. Sometimes, this facility is provided via a single switched input, or otherwise the inputs are suited only to high impedance microphones. In this case, a low impedance microphone can be made to suit by means of a matching transformer (see E&MM, December 1981).

The vast majority of vocal mixer-amplifiers use transistor circuitry, and if vocals are seen primarily as another instrument rather than as an extrinsic and straightforward means of linguistic communication, then a valve amplifier may provide more expressive results for a solo vocalist, foibles regardless. In particular, valve amplification will lessen the unpleasant consequences of grossly overloading the microphone.

Interface 1: speakers on stage

At one stroke, standing on a stage can sweep aside all the perfectionist — even romantic — aspects of playing music. In small venues, the acoustics appear to be specially designed to distort and destroy music, the stage — if it exists at all — is invariably cramped, and the distant mains socket usually has a broken switch. And even if you remembered to bring an extension cable, you still have to combat the audiences' passivity. Yet skilled musicians such as Nik Turner and Robert Fripp prefer to play small venues, however inauspicious they might seem at first sight.

The first step in accommodating the shortcomings of the small venue is to develop an awareness of acoustics. The sharp, transient sound of a handclap contains a wide range of frequencies, and can give certain useful clues as to the deadness, and the nature of any severe colouration introduced by the room. In general, long rooms with low ceilings sustain standing waves (eigentones) at low frequencies, thereby muddying the electric bass and making the bottom end of the vocal range prone to howl — (or boom!) — round. The other common hazard is an excess of absorbent surfaces — carpets, people, plants, furniture and 'acoustic' ceiling tiles. Apart from skimming off much of the high frequency content in the music, these furnishings also tend to diminish the perceived sound level overall. At the same time, the absence of equalisation will make the music seem flat and lifeless. Apart from boosting the treble, little can be done to compensate, apart from being aware of the relatively flat and demure sound as perceived by the audience, and playing a loud and strident set in defiance!

Complementary to an appraisal of the acoustics is an intuitive understanding of the sound projection properties of your speaker cabinets. Most instrument amplifiers are paired with direct radiator speakers, and apart from the tendency for the sound field to become omnidirectional at low frequencies (below circa 250Hz) and beam-like at high frequencies (above 1 to 3kHz) drivers in this configuration are mercifully quite well behaved in small venues. The vocalists' and bassist's amplification systems may use horn-loaded (i.e. bins) or vented (Thiele or reflex) enclosures however. Whilst these are superior in most respects to direct radiators, for these applications at least, they can interact in peculiar ways with the air in a small venue. Typically, the bass will be mysteriously lacking — even though the walls are shaking! A more mundane difficulty arises when a radial horn, intended to 'shoot' high frequencies over several hundred feet refuses to cover an audience at short distance. Two factors are apparent here. Firstly, horn speakers intended for large auditoriums will rarely form a coherent wavefront in the first five to twenty feet; in this region, which will often include 99% of your audience in a small club, the sound is subject to all manner of aberrations. Secondly, horn speakers are particularly adept at acting as involuntary microphones, sensing the sound energy in the room, and reflecting it back to the amplifier. This 'richochet' effect, originally proposed by Richard Elen and George Chkiantz (who also, incidentally, produced the early Hawkwind albums), can again be held responsible for all manner of spurious, if subtle, changes in the sound quality which (quote) "didn't happen at the club we were in last night".

If your vocal PA doesn't exhibit a flat response, it will invariably be necessary to place the speakers well forward of the microphone(s) and as far away in the lateral sense as possible. Careful positioning of the mics (as central as possible) and speakers (aim them away from the stage, unless there are nearby reflective surfaces — such as bare walls — in which cases 'straight ahead' will have to suffice) together with intelligent; experimentation will help to achieve usable vocal levels. If there isn't a curtain along the rear of the stage, beware of reflections arising from the sound 'out front', and experiment with the angle of the microphone. Also, beware of placing the instrument amplifiers so that they radiate directly at the vocal microphone(s); this involuntary 'miking up' will only serve to muddy the vocals and render them more unintelligible than they need be.

Once the bassist has tuned up, cabinets with castors will begin to execute a random choreography and produce peculiar percussive sounds unless they are turned over so that they lie flush with the floor... whilst discrete and lightweight bass amplifiers (or 'heads') usually benefit from a firmer or at least less agitated surface than the top of a bass cabinet; foam rubber, 'gaffer' tape, flightcase clips or even a separate amplifier stand are typical solutions here.

Interface 2: musicians on stage

The cynical quip "musicians need to hear each other slightly more than the audience needs to hear them" emphasises the crucial nature of the rapport between the members of a band playing music which is largely improvised. For instance, vocalists and lead guitarists need to hear the rhythm of the drums and bass and vice versa. Without a powerful and elaborate stage monitoring system, half the battle is to arrange yourself so you can hear the instrument(s) you need to hear most of all; deaf drummers, sitting inside one of the loudest instruments of all at the rear of the stage face the most serious and perpetual communications problem. Yet, somehow, by dint of practice and the development of intuition, they manage to play in time. And this telepathic rapport is the other half of the battle: if you cannot produce a balanced sound on stage, additional practice in working together as a unit should take precedence over thoughts about stage monitoring, unless perhaps your forte happens to be an especially loud and chaotic species of heavy metal.

The next question is "What does it sound like out front?" Of course, you can ask a friend to stand in the middle of the venue during your rehearsal, and return with comments. But unless they're perceptive, and dedicated to your long term musical interests, they're unlikely to be critical and will usually just assure you that "It sounds great". So unless you're fortunate enough to come across the services of a skilled sound engineer or fellow musician, you will have to adopt a rule of thumb: if each member of the band feels the music sounds balanced on stage then it's as near balanced out-front as makes no difference. And without a PA, your performance depends — in the end — entirely upon the discipline within yourselves; nobody outside the band can help. This factor, together with the dictates of fashion, requires that every band shall gain access to a PA willy-nilly — and the sooner the better; a giant topic to be dealt with shortly.