How do mikes work, why do they like some PAs and hate others, where do all the volts go? Andy Honeybone speaks to the world.

Bing Crosby and Billie Holiday have been credited as the first 'intimate' vocalists. Their styles were way apart but they both sang in a quiet, personal manner which was only made possible by the use of a microphone. It's easy to forget that before sound reinforcement, singing was about 'shouting the Blues' with the emphasis on projection rather than emotion.

Microphones are rather like wine in that it is difficult to say anything about a particular variety without sounding pretentious. To reduce this possibility, suitable mikes for recording the balalaika will not be proposed and only technical aspects will be discussed.

The most fundamental specification to grasp is impedance — sure, the others are important but from a practical viewpoint it could be a gig saver. Imagine you are impoverished (not difficult) and only have one mike with you at an important gig. During the sound check, the stand gets knocked to the floor and the mike develops a rattle and refuses to work as anything other than a castanet. Horror-stricken — as they say in the Tabloids — you rush off to locate the support band in the dressing-toilet and calmly ask if you could borrow a spare mike — support bands always have the best gear. To your relief the goods are handed over but there comes with it a caveat (and a holder) — 'low impedance OK?' 'Nice one' you reply, to obscure your ignorance, and off you nip to your loose-bowelled comrades. When the mike is plugged into the PA, the gain is turned up... and up... and up a bit more, but all you can hear is a faint sound amid the hiss. What's wrong? Nothing's wrong but something's missing — a transformer.

For the sake of argument, let's assume the mike in question is dynamic which means it works on the principle of a loudspeaker in reverse. Sound vibrates the cone (diaphragm) which is attached to a coil of wire which is free to move around a magnet. This movement generates a voltage in the coil which is routed to the PA for amplification. The wire which forms the coil will have a resistance of a few tens of ohms, and the voltage produced will be about a hundred or so millionths of a volt. A PA input will typically have an input 'resistance' of 50 thousand ohms and a sensitivity of several thousandths of a volt.

The 'missing' transformer is a pair of coils arranged so that a field generated by a signal in one is picked up by the other. If the number of turns in one coil is different from the other, the output will differ from the input by the same ratio. Of course, you don't get something for nothing and the total power (volts times amps) on each side of the transformer is always equal — ignoring losses. In a coconut shell then, voltage can be boosted at the expense of current and vice versa.

Many microphones incorporate an internal transformer to raise the output level to a more manageable size and as a consequence, the boosted output level appears across a much higher 'resistance' than that of the generating microphone coil.

To transfer this energy efficiently to an amplifier, the impedance of the amplifier's input should match or exceed the nominal impedance of the microphone. The sudden change from 'resistance' to impedance was deliberate and far more correct. Coils are reactive components which means their 'resistance' to alternating currents (AC) is proportional to the frequency of the signal. Impedance is the combination of pure resistive and reactive elements in a circuit.

So where does that leave our hypothetical friend back at the gig? The answer which immediately springs to mind is 'up the Swanee'. In consolation it might be worth pondering why mikes should be sold without a transformer. The answer is with the connecting cable.

Being co-axial in construction to guard the inner, signal-carrying conductor from stray mains field pick-up, the cable has an appreciable capacitance which acts as a short circuit to high frequencies. If a high impedance microphone (47K — forty seven thousand ohms) is fitted with a long lead of 20-50 feet, the high frequency sapping effect of the cable would be such that the response at 10kHz could be about ten times down on the quoted performance. If a low impedance (50ohm) mike is used instead, the cable reactance is by-passed by the very much lower signal impedance and so negligible treble is lost.

A transformer or specially matched pre-amp at the amplifier end of the cable brings the mike signal up to a usable level. Low impedance mikes are not without problems of their own as the resistance of long cable runs can be greater than the coil resistance, resulting in overall signal loss. An impedance of six hundred ohms is now favoured as a good compromise and much professional equipment is made to this standard.

Many professional mikes are 'balanced'. This means that three conductors are brought out of the microphone — one shield (earth) and both ends of the coil. The alternative unbalanced 2 wire arrangement would join one end of the coil to the shield. The advantage of the balanced system is that when the signal is converted to single-ended by a tapped transformer at the amplifier end of the cable, any grot and hum picked up by the cable is cancelled out to leave just the mike signal.

Microphones can be classed as either generators or modulators. Dynamic types are generators as they produce a voltage in response to sound pressure. Condensor or capacitor mikes require an input voltage which they then modulate. Moving coil dynamic mikes have already been explained but before capacitor types are scrutinised, consider the consequences of attaching a coil to a diaphragm. Neil would probably have a word for the arrangement, and it is found that the weight and coupling distorts the fidelity of the mike.

In a capacitor mike, the diaphragm is mechanically free and consists of a metallised plastic disc. This disc forms one plate of a capacitor, the other plate being a rigid fixture. The capacitor is charged by a polarising voltage in the region of 60-100 volts and this tensions the diaphragm. When sound hits the mike, the disc moves and the capacitance changes in sympathy with the distance between the plates. As a result, current flows in and out of the capacitor and this modulates the polarising voltage to give an output signal. Unfortunately, the impedance of the capacitor is ultra high and an internal pre-amp is needed before the signal can be let loose on the outside world.

Capacitor mikes are particularly costly because of the need for the power supply, preamp and extra connectors. Such is the price for a transducer free from colouration. Much cheaper are the Electret capacitor mikes which work on similar principles but have the high polarising voltage 'permanently' zapped within the diaphragm. Humidity (beer) causes the charge to leak away and as the diaphragm has to be made relatively thick, the frequency response is not so good. Electrets generally have a treble lift characteristic which can be used to effect when recording less than top-class cymbals. A field effect transistor (FET) pre-amplifier is usually incorporated in this class of mike and it is this part which demands the 1.5 volt battery.

The pick-up area of a mike is determined by the selective introduction of sound pressure onto the rear face of the diaphragm. If the chamber behind the diaphragm is sealed, the unit is omni-directional and picks up sound from all sides. When the chamber is vented, generally directed sound will appear on both sides of the diaphragm and so cancel out. Sounds made directly in front of the mike will hit the front of the diaphragm only and therefore be 'heard'. If a response map is drawn for this uni-directional case, the pattern is heart shaped and is given the name cardioid which comes from the Greek for the 'ticker'. Vocal mikes should be unidirectional to avoid howl-round from monitor pick-up. Omni-directional types are best left for broadcasting and bugging.

The frequency response of a mike should never be taken for granted because it can limit the maximum vocal level possible in PA applications. All mikes exhibit resonances and these show up as peaks on the frequency profile. These peaks will be the frequencies at which feedback will occur. If there is only one peak, you stand a good chance of flattening it with a parametric equaliser. Several peaks mean problems.

Radio mikes are all the rage and as a result we are suddenly faced with legalities. The Government's Radio Regulatory Division has allocated 15 frequencies in the approximate 174-175MHz range and there are tough specifications concerning stability and output power to which equipment must conform to be granted a licence. Although the licence is about £1 a year, the hardware is expensive. At the other end of the scale, Maplin Electronics sell an XLR plug adaptor which converts any mike into a transmitter for reception on a standard 88-108MHz FM stereo radio. Maplin do point out that the product is not licensable but omit to tell you that it is illegal for you to use it.

A last flurry of figures before I'm dragged off. Mike stand threads: American and Japanese — 5/8ths inch, 27 threads per inch. United Kingdom 1/2 inch, 26 threads per inch. European — 3/8ths inch, 23 threads per inch. Now you know.

Finally, if you need to give anyone a tip, tell them this. Always switch a dynamic mike off when you put it away. Although there are no batteries inside, throwing the switch shorts the coil and gives shock protection by electromagnetic damping. Any movement of the coil produces a voltage which is fed back to generate motion — like a loudspeaker — to oppose the force.