Capacitors: Audio signals

In the previous issue, after two chapters devoted to power supplies, we saw how the characteristics of capacitors are exploited in that part of a component, using them as a filter (to obtain purer direct current) and also as an energy store (so that the power supply can cope with transient current demands).

Remember that a capacitor is a passive electronic component that stores electric charge and presents infinite resistance to direct current; in other words, direct current will never flow through a capacitor except for a brief instant while it charges.

DC BLOCKING

This characteristic already gives us a first use for capacitors: blocking direct current when it may accompany an audio signal (which, electrically speaking, is an alternating current). This is precisely how almost all tube preamplifiers and sources use them at their output, since there is a very considerable amount of direct current at the output of the circuit which, if it passed to the next circuit (a power amplifier), could damage it. These are the "coupling" capacitors, located in the output section of these components, and they are usually quite large, since the greater the capacitance, the less they affect the bass frequency response (we will see why shortly).

For the same reason and as a precaution, some power amplifiers and preamplifiers also have capacitors in line with the incoming signal to eliminate any residual direct current that may be present. What's more, in components from industrial (rather than artisan) manufacturers, it is common to find coupling capacitors between the various internal sections of the component, so that if one of them develops a fault and passes direct current, it does not affect the next link in the chain.

As you can easily imagine, every time the audio signal passes through a capacitor, especially if (as in the last case mentioned) it is not of optimal quality, it can be degraded. Hence the interest in minimalist components, and the reason why some brands even use this highly technical issue in the advertising for their products: "no capacitors in the signal path", or "with just one high-quality capacitor in the audio signal path".

For this purpose, the worst possible capacitor is an electrolytic but, precisely because this type of capacitor offers high capacitance in a compact size (and at a low price), it is the most commonly used in mass-market or budget components.

To avoid using capacitors in this way, the circuit can be designed so that it is intrinsically impossible for it to output direct current; alternatively, the output capacitors can be replaced by a circuit that detects any small amount of direct current and cancels it: these are known as DC servo circuits.

PASSIVE FILTERS

We saw how capacitors could filter the current from a power supply so that it is as pure as possible. Here we are going to look at another use in another type of filter: removing a certain range of frequencies. We have already seen how useful they are for filtering out direct current, which is nothing other than an audio signal at 0 Hz.

A good example is the crossover of a two-way loudspeaker, in which, from the full signal coming from the amplifier, we want only the bass to go to the large driver, optimized for it, and the treble to go to the tweeter, which in turn is optimized for high frequencies. If a tweeter receives bass, it will be unable to reproduce it correctly and will produce distortion, and the same happens with a bass driver fed treble. The crossover takes care of precisely that, dividing the frequencies between the two drivers. And the simplest possible filter is precisely a capacitor: placed in series with the tweeter signal, it will remove the bass. The frequency above which it acts is calculated with a formula that need not concern us here and which depends on the resistance of the circuit, but bear in mind that the lower the capacitance, the higher the frequency, and vice versa; in other words, the greater the capacitance, the lower the cutoff frequency (a cutoff that in this case is not steep, quite the opposite, since this is a first-order filter, with only 6 dB of attenuation per octave).

This explains the use of high-capacitance capacitors to filter DC from the audio signal: if we used a smaller one, the cutoff frequency could fall within the audible range and leave the component with a curtailed bass response. An example: a 10-microfarad capacitor at the output of a preamplifier will form a filter with the following stage at 5 Hz, which is inaudible; but if that capacitor had a thousand times less capacitance, the cutoff would be at 40 Hz, and then we would immediately notice the missing bass.

A capacitor can also be used to filter out treble, in this case connected in parallel with the component we want to filter. Again, the cutoff frequency will change depending on the capacitance.

Hardly any loudspeakers use such simple filters, since manufacturers make better use of the qualities of their drivers with steeper or more complex filters. But these are always combinations of capacitors, resistors and inductors, and the operating principle is exactly the same, since all these components are passive, hence the name of this type of filter.

CAPACITORS FOR AUDIO

Last month we saw that a very common type of capacitor is the electrolytic, widely used in power supplies because of its high capacitance (and therefore good filtering capability) and abundant energy storage. However, when an audio signal passes through a capacitor of this type, it is easy to measure distortion and hear the sound quality deteriorate. From a quality standpoint, to allow audio signals to pass with minimal degradation, these are the types of capacitors available (in order of quality) according to the dielectric used inside them: Teflon (PTFE), polystyrene, metallized polypropylene, polypropylene, metallized polycarbonate, polycarbonate, metallized polyethylene, polyethylene, mica. The first (Teflon) are very expensive and are usually too large to be practical, and for both reasons they are only found in very high-end components. There are also oil-impregnated paper capacitors, which would rank among the best, but once again price and size make them impractical. That is why the most common types in quality mid-range and high-end components are polystyrene, polypropylene and metallized polyester (the latter two also known by their abbreviations MKP and MKT respectively).

There are also some types of electrolytics that, thanks to their construction and the materials used, are considerably better than ordinary ones. Notable examples are Black Gate, Nichicon Muse and Elna Silmic.

SUMS AND SYNERGIES

In the previous chapter we saw that by combining capacitors we obtain another capacitor whose capacitance is the sum of them all (if they are in parallel). In some audio applications, the final value matters less than the properties. That is why several types of capacitor are sometimes used in parallel, so that each contributes its best qualities. A very common example is to use, as the decoupling capacitor at the output of a preamplifier, a fairly high-capacitance polypropylene capacitor (thus ensuring good bass) together with a much lower-capacitance Teflon one, which will help in the mid and high frequencies.