Transformer coupled class B, max efficiency, crossover

Circuit diagram

       T1 (input)                                   T2 (output)
                        +--- B1  Q1 (NPN)  C1 ---+
                        |         E1             |
   o----+  ||  +--------+         |              +--------+  ||  +----+
        )  ||  (                  |                       (  ||  )    |
  Vi    )  ||  (-- CT -- GND     GND          VCC -- CT --(  ||  )   [RL]
        )  ||  (                  |                       (  ||  )    |
   o----+  ||  +--------+         |              +--------+  ||  +----+
                        |         E2             |
                        +--- B2  Q2 (NPN)  C2 ---+

The centre-tapped input transformer T1 gives two drives of opposite phase to the bases of Q1 and Q2; both emitters are grounded. Q1 conducts on one half-cycle and Q2 on the other. The centre tap of the output transformer's primary is connected to VCC, and the collectors go to its two ends. The two half-cycle currents combine in the secondary to give the full waveform across RL.

Maximum efficiency

Let the peak collector current be Im and the peak voltage across one half of the primary be Vm = Im R'L, where R'L = (N1/N2)² RL is the load reflected into one half of the primary (N1 = turns on each half of the primary, N2 = secondary turns).

• Output power: Po = Im² R'L/2
• Input power: Pi = VCC · (2Im/π) (two half-sine current pulses)

η = Po/Pi = (π/4)(Im R'L/VCC) = (π/4)(Vm/VCC)

The maximum swing is Vm = VCC, so

ηmax = π/4 = 78.5 %

Crossover distortion

Crossover distortion is the distortion at the zero crossing of the output of a class B stage. For inputs between about −0.7 V and +0.7 V, neither transistor is forward biased, so the output stays at zero (a dead zone) and the waveform is flattened where it crosses zero. The effect is most visible at low signal levels. It is cured by biasing the transistors slightly above cut-off (class AB).