[Part 1 offers an overview and introduction to the sources of distortion in power amplifiers.]
The input stage of an amplifier performs the critical duty of subtracting the feedback signal from the input, to generate the error signal that drives the output. It is almost invariably a differential transconductance stage; a voltage-difference input results in a current output that is essentially insensitive to the voltage at the output port. Its design is also frequently neglected, as it is assumed that the signals involved must be small, and that its linearity can therefore be taken lightly compared with that of the voltage amplifier stage (VAS) or the output stage. This is quite wrong, for a misconceived or even mildly wayward input stage can easily dominate HF distortion performance.
The input transconductance is one of the two parameters setting HF open-loop (o/l) gain, and thus has a powerful influence on stability and transient behaviour as well as distortion. Ideally the designer should set out with some notion of how much o/l gain at 20 kHz will be safe when driving worst-case reactive loads " a precise measurement method of open-loop gain was outlined last month " and from this a suitable combination of input transconductance and dominant-pole Miller capacitance can be chosen.
Many of the performance graphs shown here are taken from a model (small-signal stages only) amplifier with a Class-A emitter-follower output, at +16dBu on 15V rails. However, since the output from the input pair is in current form, the rail voltage in itself has no significant effect on the linearity of the input stage. It is the current swing at its output that is the crucial factor.
Vive la differential
The primary motivation for using a differential pair as the input stage of an amplifier is usually its low DC offset. Apart from its inherently lower offset due to the cancellation of the Vbe voltages, it has the added advantage that its standing current does not have to flow through the feedback network.
However a second powerful reason is that its linearity is far superior to single-transistor input stages. Figure 1 shows three versions, in increasing order of sophistication. The resistor-tail version in Figure 1(a) has poor CMRR and PSRR and is generally a false economy; it will not be further considered. The mirrored version in Figure 1(c) has the best balance, as well as twice the transconductance of that in Figure 1(b).
Figure 1: Three versions of an input pair: (a) Simple tail resistor; (b) Tail current-source; (c) With collector current-mirror to give inherently good Ic balance.
Intuitively, the input stage should generate a minimal proportion of the overall distortion because the voltage signals it handles are very small, appearing as they do upstream of the VAS that provides almost all the voltage gain. However, above the first pole frequency P1, the current required to drive Cdom dominates the proceedings, and this remorselessly doubles with each octave, thus:
Ipk = 2pF • Cdom • Vpk
For example the current required at 100 W, 8 O and 20 kHz, with a 100 pF Cdom is 0.5 mA peak, which may be a large proportion of the input standing current, and so the linearity of transconductance for large current excursions will be of the first importance if we want low distortion at high frequencies.
Figure 2, curve A, shows the distortion plot for a model amplifier (at +16dBu output) designed so that the distortion from all other sources is negligible compared with that from the carefully balanced input stage. With a small-signal class-A stage this essentially reduces to making sure that the VAS is properly linearised. Plots are shown for both 80 kHz and 500 kHz measurement bandwidths to show both HF behaviour and LF distortion. It demonstrates that the distortion is below the noise floor until 10 kHz, when it emerges and heaves upwards at a precipitous 18 dB/octave.
Figure 2: Distortion performance of model amplifier differential pair at A compared with singleton input at B. The singleton generates copious second-harmonic distortion.
This rapid increase is due to the input stage signal current doubling with every octave to drive Cdom; this means that the associated third harmonic distortion will quadruple with every octave increase. Simultaneously the overall NFB available to linearise this distortion is falling at 6 dB/octave since we are almost certainly above the dominant pole frequency P1. The combined effect is an 18 dB/octave rise. If the VAS or the output stage were generating distortion, this would be rising at only 6 dB/octave and would look quite different on the plot.
This form of non-linearity, which depends on the rate-of-change of the output voltage, is the nearest thing to what we normally call TID, an acronym that now seems to be falling out of fashion. Slew-induced distortion SID is a better description of the effect.
If the input pair is not accurately balanced, then the situation is more complex. Second as well as third harmonic distortion is now generated, and by the same reasoning this has a slope of closer to 12 dB/octave. This vital point requires examination.