|
So i got tired of class A amps and decided to try a class AB topology for some of its superior characteristics. It turned out to be a very good example with THD levels rivalling even some of the best class A topologies. Some of its specifications are:
- Output Power - 75W (rms)
- Drawn Power - 108W
- Efficiency - 70%
- Frequency Response - 4Hz - 20Khz @ -3db
- Input sensitivity for full output - 500mV and that value can be brought down with feedback adjustment.
- THD+N - Extremely low with no prominent switching point noise. Measured it 0.008% at 1Khz and full output power, with almost no zero point switching noise. In reality it would be more like 0.05% which is also remarkable for a simple topology as this.
- Ripple rejection - The external circuitry has a supposed ripple rejection of -350dB. That's extremely high, but is quite dependent on the quality of the op amp, its ripple rejection and the regulation of the op amp supply. In reality a probable value would be around -100db provided the certain needs. At -100dB a 1V ripple would be rejected down to 10uV
Some remarks on the schematic:
- The 2 by 2 diodes are actually 4 in reality. I modified these in EC to save some space.
- The output transistors must be a parallel pair of 2N3055 and its MJ2955 PNP equivalent. If you try to use only 2 of them and not the full set of 4, you'll end up almost reaching their maximum power rating, at which transistors tend to burn out quickly.
- The op amp supply rail should be absolutely regulated for the amp to respond to the given characteristics. The regulation is done not with a Zener diode, but with an RC filtering network biasing a transistor. If left unattended amplifier may severely deviate from the given characteristics (especially THD ripple rejection etc.) Worst case scenario, the amp starts oscillating uncontrollably which may ultimately lead to the destruction of the power transistors.
|
