|
These are three of the most basic phase splitter topologies. The first one is probably the easiest to grasp. It's sometimes called a split load phase inverter, because the collector load resistor and the emitter load resistor are equal. I personally prefer it, for it's high linearity. It operates at 50% internal negative feedback, so it's considerably linear, before it is overdriven, at which point a very gross distortion appears, where the wave is doubled, as it starts acting as a full wave rectifier. One of the problems with it is that the output impedances of the two outputs are not equal. The output impedance of the collector output is roughly equal to the collector resistor and the output impedance of the emitter output is calculated the same way an emitter follower impedance is. This somewhat makes it hard to balance when loading the two outputs, but careful consideration makes it a perfect phase splitter. The advantages are simplicity and linearity, disadvantages are tendency to gross distortion and sometimes problems to balance the outputs. The second circuit is a differetial pair, which is basically a differential amplifier, with emitter connection. One end works in a common emitter mode, and the second arm works in a common base mode. Perfect symmetry can be obtained when the emitter load is substituted with a constant current source. The advantages of this topology, is that it doesn't need a driving voltage amplification stage, as it has some voltage gain. The other advantage is that when done properly it's very self regulatory and balanced. I've seen many professional circuits use such a topology. From all three presented here, it has the most potential for best balancing. The disadvantages are that it uses 2 active devices to split the phase and is generally much more complicated than a split load inverter topology. The third circuit is a slight variation of the second circuit. The outputs are again taken from the same places at the two collectors, but the second arm still operates at a common emitter mode. To balance the signals, the input signal for the second stage must be taken from the output of the first stage, but first it needs to be attenuated. To do so, you connect the two outputs in a single point through two resistors, so that the opposing currents will cancel out to a given extend. The advantages of this topology is that it gives a very good symmetry, equal output impedances for the loads and it has a voltage gain. The disadvantages are complexity and the requirement for two active devices.
|
