Class AB amplifier
Class AB
Class AB operation falls between Class A and Class B . For example,
the push-pull concept (with improved efficiency as compared to a Class A architecture)
can be used, while at the same time improving the linearity (as compared to a Class
B architecture) by minimizing crossover distortion. , the
conduction angle ? is greater than ? (pure Class B) but less than 2? (pure Class
A). Thus each device conducts for a slightly longer period of time, which smooths
out the output voltage waveform during the crossover period. This approach can
therfore provide linearity closer to class A and efficiency closer to class B. Depending
on the linearity and efficiency requirements, the designer can choose the bias point
appropriately (closer to threshold for higher efficiency or closer to the centre of the
load line for better linearity). RF PAs in Class AB architectures have been widely
reported in the literature, with efficiencies ranging between 30% to 60%
Class C
In a Class B amplifier, each transistor is biased at threshold and conducts for half a
cycle. In a classical class C amplifier, the transistor is biased below threshold so that the transistor is active for less than half of the RF cycle. Because
of this biasing strategy, the efficiency is increased, since the device is on for less than
half of the cycle [the conduction angle is less than ? , but the linearity
is significantly degraded. The efficiency can be increased theoretically to a maximum
of 100% if the conduction angle goes to 0?. Of course this causes the output power
to decrease to 0 since no current is being delivered to the load. A compromise of,
for example, ? = 150? can result in a PAE of 85% . Such a PA can be used
in applications with constant envelope modulation schemes where linearity is not a
major concern.